External link

External link
Understanding Human Factors
a guide for the railway industry
Understanding Human Factors/June 08
About this document
In an industry with multiple stakeholders, Rail Safety and Standards
Board (RSSB) builds consensus and facilitates the resolution of
difficult cross-industry issues. RSSB provides knowledge, analysis, a
substantial level of technical expertise and powerful information
and risk management tools and delivers this unique mix to the
industry across a whole range of subject areas. Working with our
industry partners our purpose is, therefore, to:
The scope of this human factors guide
• continuously improve the level of safety in the rail industry;
• drive out unnecessary cost; and
• improve business performance.
RSSB continues to manage an extensive programme of research
and development on the industry’s behalf. This work addresses
a wide range of operational and business issues of common
concern to Members. Its outputs feed into collective industry
planning, the formulation of safety standards and safety decisionmaking from individual Members.
How was this Guide developed?
This guide is the culmination of a process that started in 2003,
when the Health and Safety Executive (HSE) and RSSB’s Human
Factors Team identified the need for a comprehensive human
factors guide for the railway industry in Great Britain. Industry
stakeholders endorsed the development of the Guide in January
2004, when they identified it as a priority project for RSSB.
In developing the Guide, RSSB adopted a user-centred design
approach and interviewed a representative sample of the Guide’s
end users early in the project. The results of these interviews
were used to inform the content and style of the Guide. End
user representatives, as well as the Railway Industry Advisory
Committee (RIAC) Human Factors Working Group and RSSB’s
technical advisers, reviewed the Guide at various stages of its
development.
This Guideis available in both print and searchable, hyperlinked pdf
formats.
Understanding Human Factors/June 08
Page i
Contents
Who is this Guide for?
Part 1: Getting started
3 Design
This Guide has been created for designers, suppliers,
managers, supervisors, trainers and Health & Safety (H&S)
staff who work in the railway industry. Its main purpose is
to answer the question ‘what practical advice can a human
factors approach offer to railway staff without requiring them
to be experts in the subject?’ Accordingly, this Guide is not
written as a text book, or a step-by-step process. Instead,
it uses a Frequently Asked Questions (FAQ) format,
addressing questions that are relevant to railway staff, and
which human factors research has had some success in
answering. The answers we give are in sufficient detail to
tell you what a human factors approach involves, why it’s
important and what you can do. The questions are not
presented in any particular order. Each one represents a
potential channel of communication between rail industry
and human factors professionals. The Guide will have met
an important goal if it encourages railway professionals
to seek more frequent, more timely, and more informed
dialogues with their human factors colleagues. This Guide
will also be of interest to front-line staff who would like
to know how some of the decisions that affect their
working lives should be made.
What are human factors?
Where are human factors important?
What is the organisational payoff?
How should I navigate this Guide?
3
3
4
5
Part 2: Guidance
9
Human performance
9
How should you use this Guide?
This Guide has been made for browsing. The best way
of using it is to scan the questions covered and dip into
those which seem of particular interest to you. We hope
you are rewarded by insight, information and signposts to
further sources of advice. If this Guide causes you to ask
questions that you would not otherwise have asked, it will
have done its job.
Page ii
Why do people make mistakes?
Why do people break rules?
Why do people take risks?
Why do accidents happen?
9
12
15
18
25
User-centred design
25
Why is user-centred design important?
What is required to do user-centred design?
How is user-centred design done?
How does user-centred design fit with systems
engineering?
What is a user requirement?
How can usability be tested?
25
25
27
Equipment design
31
What makes good equipment?
What are the principles for equipment design?
What makes controls easy to operate?
What makes interfaces easy to use?
31
31
32
33
Function allocation
37
What is function allocation?
How is function allocation done?
When is function allocation done?
37
37
40
Workplace design
41
How important is it?
What are the principles of workplace design?
How is workplace design best approached?
What should be taken into account?
41
41
42
42
Task analysis
47
What is task analysis?
How is task analysis done?
What is Cognitive Task Analysis?
47
48
49
Job design
50
What is job design?
What should be taken into account?
50
50
27
27
28
Understanding Human Factors/June 08
Contents
Training
55
Culture
91
Conditions
117
Training needs analysis
55
Leadership
96
Morale and motivation
117
What is training needs analysis?
When is training needs analysis done?
How is training needs analysis done?
55
55
56
117
118
119
61
96
96
96
97
What are morale and motivation?
What motivates people at work?
How can you improve morale and motivation?
Cost-effective training
What is leadership?
What style of leadership works best?
What should leaders do?
How can leaders bring about a safety culture?
Stress
120
What is training?
How do people learn?
How should people be trained?
How can skill fade be prevented?
How should training be evaluated?
61
61
62
64
65
Management 99
What does a manager do?
What makes a good manager?
How can you become a better manager?
How can managers bring about a safety culture?
99
99
101
101
What is stress?
What are the effects of stress?
What causes stress?
How can you recognise stress?
How can you reduce stress?
120
121
121
122
123
Supervision and appraisal
68
Teamworking 103
Workload
125
What is a team?
What makes a good team?
What is teamwork good practice?
How does teamwork affect safety culture?
How can you improve teamworking?
103
103
104
104
105
What is workload?
How much workload is too much?
How do you measure workload?
How do you identify workload problems?
125
125
126
127
Shift work
128
Communication
106
Why is communication so important?
Why is communication so difficult? How do communications help with safety culture?
How can you improve communications?
How can you manage communications?
106
106
107
108
108
What’s the problem with shift work?
How can you reduce shift work problems?
128
129
Change
109
What drives change?
What are the obstacles to change?
How can change be managed?
Can safety culture be changed?
What are the principles of change management?
109
109
110
111
113
What should a supervisor actually do?
68
What’s the difference between competence assessment
and performance appraisal?
69
How do you build an effective team?
71
How do you handle a difficult conversation?
73
How do you negotiate effectively?
74
Staffing
79
Selection
79
How do you select the right people?
Which selection method is best?
How should you conduct a selection interview?
79
81
82
Recruitment
84
How can you recruit the right people?
How many people is enough?
84
86
Retention
87
Why is retention important?
How do you keep the right people?
87
87
Understanding Human Factors/June 08
Page iii
Contents
Part 3: Reference
135
Techniques
135
Applied Cognitive Task Analysis
Brainstorming
Checklists
Cognitive Mapping
Cognitive Walk-through
Communications Usage Diagram
Critical Decision Method
Critical Incident Technique
Design Scenario Analysis
Fault Trees
Focus Groups
Groupware Task Analysis
Heuristic Analysis
Hexagons
Hierarchical Task Analysis
Human Error Assessment and Reduction Technique
Human Error HAZOP
Human Reliability Analysis
Instantaneous Self-Assessment (of workload)
Integrated Performance Modelling Environment
Interface Surveys
Interviews
Keystroke Level Model
Layout Analysis
Link Analysis
Murphy Diagrams
NASA-Task Load Index
Observational Analysis
Questionnaire
139
140
141
142
143
143
144
145
145
146
147
147
148
149
149
151
152
153
153
154
155
155
156
157
157
158
159
159
160
Page iv
Questionnaires for Distributed Assessment of Team
Mutual Awareness
Questionnaire for User Interface Satisfaction
Sequentially Timed Events Plotting
Situation Awareness for SHAPE
Situation Awareness Rating Technique
Soft Systems Methodology
Software Usability Measurement Inventory
Subjective Workload Assessment Technique
System Usability Scale
Systematic Human Error Reduction and Prediction
Approach
Task and Training Requirements Methodology
Task-Centred System Design
Team Cognitive Task Analysis
Team Decision Requirements Exercise
Team Workload Assessment
Teamworking Improvement Process
User Trial
Walk-through Analysis
Why-Because Analysis
Workshops
161
161
162
163
164
164
165
166
167
167
168
169
170
171
171
172
173
173
174
175
Jargon buster
179
Bibliography
189
Index
207
Understanding Human Factors/June 08
understanding
Getting started
human factors
understanding
Getting started
human factors
Part 1
Getting started
Part 1: Getting started
What are human factors?
Human factors is another term for ergonomics. It has
traditionally focused on ensuring that employees have
safe and easy-to-use equipment and a place in which they
can work efficiently. However, we use the term ‘human
factors’ in this Guide in a much broader sense.
each other. The whole of the railway industry will only
operate at its best if it attends to all the human factors
that can affect its performance – that is, its safety and
profitability.
The critical areas of human factors
In particular, the timely application of human factors
knowledge and techniques:
• reduces the potential for error
This broader focus is necessary because of several
interconnected trends:
• Technical systems are becoming more wide-reaching
and complex, making it necessary to consider their
effect on the larger work group and, indeed, the total
organisation
• Work is placing increased demands on people’s
knowledge
• Organisations are increasingly regarding employees
– as well as technology – as valuable investments.
Instead of just focusing on the relationship between the
individual and their equipment and working environment,
we need to ensure there is a good balance between the
organisation as a whole, its people, working practices
and technology. Consequently, this Guide is based on a
definition of human factors as:
‘all the ‘people’ issues we need to consider to assure
the lifelong safety and effectiveness of a system or
organisation.’
If an organisation attends successfully to all human factors,
the organisation and its people will get the best out of
Understanding Human Factors/June 08
• increases the margin for safety
• reduces the potential for expensive re-design
• increases the efficiency and effectiveness of training
• reduces the potential for expensive staff turnover
• increases the productivity of the whole organisation.
Where are human factors important?
The diagram on this page shows the five areas of human
factors that are critical to human performance in the
railway industry.
Design
Everything manufactured, supplied and brought into
service in the railway industry has been designed. The
products of design must be fit for purpose. Here, we use
‘product’ quite widely to include equipment, processes,
procedures, rules and rule books etc. Fitness for purpose
means adequately defining at the design stage both the
way products are to be used and the technical problems
they are intended to solve. The usability of a product
depends on several important human factors issues,
including:
• how easy it is for the user to understand the function
of the product
• how straightforward and safe the product is to
operate
• how well the product supports the user’s task
• how well the use of the product fits with related
organisational products and their users.
Ignoring the human factors of design does not simply
miss a major performance improvement opportunity: it
can lead to a severe decline in performance, loss of staff
or customers through accident or wastage, and often
substantial financial cost (eg failed IT systems).
Page 3
Part 1
Getting started
Training
Conditions
Performance
People need to be developed in ways that fulfil their
own potential as well as the needs of the organisation for
which they work. Training should be seen as a continuous
process by which organisations get the most out of
people – and vice versa. In order to be cost effective,
organisations need to gauge the right time to train the
right skills in the right people in the right way.
Of course, standard conditions of work are usually
defined in appropriate detail by legislation and
employment contracts. The human factors approach
is concerned with the impact of workload, shift work,
morale, motivation and stress on performance and wellbeing.
Within an organisation, human performance is directed
behaviour that takes place for some measurable purpose.
What is the organisational payoff?
Staffing
Organisations need the right numbers of people doing
the right jobs at the right time – and once they’ve
got them, they need to keep hold of them. Recruiting,
selecting and retaining the right people are all crucial to
the success of an organisation. Getting it wrong is very
expensive and can lead to organisational weakness and
even collapse.
Culture
Each organisation automatically develops its own culture.
Culture is both a product and a cause of the way people
behave with each other. An organisation’s culture is
apparent in the behaviour of its leaders, its teams and its
managers, and in the style and expectations with which its
people communicate with each other. It’s also responsible
for how easily, or not, an organisation can change. An
organisation can’t change culture directly. But it can find
ways of influencing people’s behaviour – which then
influences its culture and (hopefully!) encourages more of
the desired behaviour.
Page 4
When a safety-critical organisation works on human
factors in all five critical areas described above, a new
possibility emerges – that of becoming a high reliability
organisation (HRO). An HRO is one that has learned
to manage the unexpected by being chronically uneasy.
It grows suspicious ‘if things get too quiet around here’.
As a result, it is able to notice the unexpected in the
making and then halt – or at the very least, contain – its
development. And if some of the unexpected breaks
through the containment, the HRO focuses on keeping
the errors small and quickly getting the system to
function again.
You can find out more about HROs on page 95 in the
section on Culture. Striving towards being an HRO
means addressing
human factors in
‘If eternal vigilance is
all five critical areas
the price of liberty,
described above.
These five areas
then chronic unease
are important
the price of safety.’
because they all
James Reason, psychologist
influence human
performance.
is
How well this purpose is achieved will depend on both
external and internal conditions. By external conditions
we mean environmental, cultural and organisational
factors that affect people’s behaviour from outside. By
internal conditions we mean the set of psychological,
physiological and anatomical factors that shape people’s
behaviour from within. Many external conditions, such as
ambient temperature and noise, can be controlled. Some
internal conditions, such as muscular power and the
trainability of individuals, must be taken as ‘givens’.
However, there are some enduring features of the
nature of human performance that are very difficult for
managers, designers and trainers to deal with. People at
work make mistakes. They take risks. They break rules.
They have accidents.
In any particular organisation, it may be appropriate
to address these fundamental aspects of human
performance by any arbitrary mix of the five critical areas
– design, selection, training, culture or conditions. While
we cannot offer guidance about the best mix for your
organisation – since each context will be different – we
can offer some insight into the nature of these aspects of
human performance.
This we do in the first section of Part 2.
Understanding Human Factors/June 08
Part 1
Getting started
How should I navigate this Guide?
Each of the sections in Part 2 begins with a wheel
diagram, based on the main wheel diagram at the
beginning of this Guide. Each wheel diagram identifies the
relevant human factors topics and questions dealt with
in the section it introduces, together with relevant page
numbers. The topics are illuminated by examples, case
studies and other explanatory panels.
Text that appears in italicised blue signifies the title of
other topics in Part 2 of the document. Text that appears
in italicised green signifies a technique for which more
detail is provided in Part 3. At the end of each section,
you will find a set of selected references that will be
useful if you wish to delve further into the topics raised.
In Part 3 you will find an extensive bibliography, a jargon
buster section and a set of representative human factors
techniques. Looking through them will give you a good
sense of what is involved in collecting and analysing
human factors information.
If you are using the on-screen (pdf) version of the Guide,
you will find that the coloured text and page numbers
are also links which you can click for fast access to other
parts of the document. In addition, website references
and red text (ie references) are also clickable links which
will access the Internet if you are connected. For best
results, please familiarise yourself with the navigational
controls provided by Acrobat Reader. These controls will
also allow you to quickly navigate around the document,
including going back to the page you just came from.
Understanding Human Factors/June 08
Page 5
understanding
Performance
human factors
understanding
Performance
human factors
Part 2: Guidance
Human performance
Part 2: Guidance
Human performance
The diagram below focuses on what
is at the core of human factors
– human performance. It focuses
on four fundamental questions
about humans at work. This part of
the Guide deals with each of these
questions in turn. At the end of each
question you will find a list of sources
of further information that will
provide more detail.
Questions of performance
Why do people make mistakes?
What sorts of mistakes do people make?
There are three types of performance that can lead to
People get things wrong all the time. Even the most
errors:
straightforward task, such as reading numbers from a
display, is not error-free. People simply
performance – where we routinely perform
‘An expert is a person must make mistakes. It is a crucial part • Skill-based
highly practised activities with little conscious effort, eg
of our nature. A mistake is something
who has made all the
replacing a rail clip or setting a route on an NX panel
that gives us information about
mistakes that can be
how near our goal we are. Without
• Rule-based performance – where we have more mental
involvement and apply previously learned rules to tasks
made in a very narrow mistakes there can be no learning.
Errors are at the heart of our ability
we have usually been trained for, eg planning a route in
field.’ Niels Bohr, physicist
to adapt to, and master, new situations.
a signal box before setting it, talking a driver past a red
On the other hand, some errors have
aspect, or evacuating a station
such dire consequences compared with their learning
• Knowledge-based performance – where we have even
value that they simply need to be prevented wherever
more mental involvement, often in novel situations,
possible. In the 1889 Armagh train disaster, 88 people
eg attending an accident scene, or counselling a staff
were killed, most of them children, when their overloaded
member.
train stopped, was divided by railway staff, and then rolled
back down a hill colliding with the train behind. In those
Errors in skill-based performance
days, signallers separated trains by the clock: they sent
Errors in skill-based performance are generally regarded
trains into the next section on the basis of how long
as slips or lapses. They typically occur when our attention
it had been since they sent the last one. After Armagh,
is diverted and we fail to monitor our actions. There are
interlocking and absolute block working were made
several types:
compulsory. In addition, automatic
‘Making mistakes
brakes were fitted so that split trains
• Familiarity slips – where something we
could not lead to runaways.
frequently do ‘takes-over’ a similar but less
simply means you
are learning faster.’
familiar action. For example, we might dial
This accident was not due to a mistake
a frequently used telephone number when
Weston H. Agor, Professor of
by the signaller, or anyone else on the
intending to dial a similar one
Public Administration
day. Rather, it was a consequence of
• Similarity slips – where the intended
the design decisions taken by people
action is similar to other actions we do a lot, so that
who had not anticipated the particular combination of
we perform the right action on the wrong object. For
circumstances that occurred at Armagh. See Why do
example, a signaller might normalise the wrong points
accidents happen? on page 18 for more about the real
switch on a panel because it is close to similar switches
relationship between accidents and human error.
Understanding Human Factors/June 08
Page 9
Part 2: Guidance
Human performance
• Memory lapses – where we forget the goal in the
middle of a sequence of actions or omit a step in a
routine sequence. An example is forgetting why we
entered a particular room. Another is the driver who
forgets to use their AWS ‘sunflower’ display to remind
them that they have overridden a warning to slow
down. Yet another is the signaller who is distracted
during arrangements for a possession and fails to use a
reminder appliance as required
AWS design – a problem of association
The Automatic Warning System installed on all
passenger trains in the UK is an example of a system
that was not designed with limitations of human
attention in mind. It is a device fitted in the train cab,
based on the now obsolete mechanical system of
signalling that used to signal either STOP or PROCEED.
It sounds a bell when a clear (green) signal is passed
and a buzzer when caution or danger is signalled. The
AWS is a useful safety system in that if the buzzer is
not acknowledged by the press of a button, then the
train begins to stop automatically. However, times have
changed since it was designed. In today’s commuter
traffic, most signals will be at the ‘caution’ aspect, and
given the frequency of signals (spaced 1km apart), most
drivers will face two signals per minute. Since people
‘automate’ highly repetitive behaviour, drivers can lose
focus on the reasons for carrying out this repetitive
task, and act in reflex whenever the buzzer sounds. The
end result is that drivers often hear the buzzer and
press the button reflexively without thinking about train
speed and location.
Source: Davies (2000), reproduced with permission from POSTNOTE, Jun
2001, Parliamentary Office of Sci and Tech.
Page 10
• Association slips – where the brain makes a faulty
• Confirmation bias – looking for information that
connection between two ideas, often when one is
confirms belief about the situation, while ignoring or
an external stimulus that typically provokes a certain
filtering out anything which disagrees. This can arise
action. An example is a driver reacting to one alarm as
through over-confidence, or as a way of making shortif another was going off. Another common driver error
cuts and reducing complexity
arises from the design of the AWS
(see Panel, AWS design – a problem ‘If everything seems to
Either of these biases can lead us
of association).
into making faulty conclusions about
be going according to
a situation, and so drawing up and
executing a faulty plan to accomplish
Errors in rule-based performance
plan, you’ve obviously
the task.
These are mistakes we make in
applying known rules. For example:
overlooked something.’
Commercial Pilot
• Misapplying a good rule – that
is, applying the rule in a situation where it is not
appropriate. This is often a rule that is frequently used
and seems to fit the situation well enough
• Applying a bad rule – so that in certain situations the
job gets done, but with unwanted consequences
• Failing to apply a good rule – that is, ignoring a rule that
is applicable and valid in a certain situation.
Errors in knowledge-based performance
Knowledge-based performance is especially prone to
errors, often of a complex nature. They arise from a
lack of knowledge, uncertainty, lack of concentration,
or a misapplication of knowledge, particularly in novel
situations. Examples are:
• Availability bias – choosing a course of action because
it is the one that comes most readily to mind. For
example, the Armagh railway disaster was a result of a
train movements procedure that was straightforward
to implement, but fundamentally flawed
What causes errors?
Errors are not random events. Rather, they are a
consequence of what normally goes on in our mind,
arising because of inattention, incomplete knowledge,
sparse sensory data, mis-perceptions, forgetting
something, problems in our relationships with colleagues,
friends and family, and so on (see Panel, Errors are
consequences – not causes). In turn, many of these factors
Errors are consequences – not causes
A major study of over 100 rail accidents and near
accidents over three years in Australia found that driver
inattention was the most important factor – especially
inattentiveness to railway signals. Significantly, the
report concluded that ‘the problem of sustained
attention amongst drivers rests with the higher levels of
the organisation where work conditions are designed and
controlled.’ In other words, the real cause of many of
these accidents was not driver error, but the conditions
in which drivers had to work. Driver error was seen to
be a consequence of the problem, not the cause of it.
Summarised from research reported in Edkins and Pollock (1997)
Understanding Human Factors/June 08
Part 2: Guidance
Human performance
are shaped by the operational context in which we work,
including the social climate, the management culture, our
working conditions and the fitness for purpose of the
tools we work with.
the problem to one of how to create the best possible
interface between people and the automated component.
At worst, it creates even more problems when the
automation fails in front of a bored, de-skilled user. See
function allocation on page 37.
‘Reducing human error
One of the key causes of human
error is having our attention
involves activity in every Training – is used to ensure that
people are rehearsed in their skills
diverted. Attention can be diverted
area
of
human
factors
and knowledge and therefore less
by environmental factors, such as
likely to make mistakes, and are
alarms, telephone conversations
just
tackling
one
is
never
better able to recover from mistakes
and demands by other people; by
enough.’
when they do make them. However,
physiological factors, such as fatigue,
thought needs to be given to the
sleep loss, alcohol, drugs and illness;
and psychological factors, such as having to juggle multiple type of training required. More training on skills, rules
and knowledge is of little benefit to people who commit
activities, stress, boredom, frustration, fear, anxiety, anger
deliberate violations. See rule-breaking (page 12) and
and personal worries.
risk taking (page 15). Violations are better dealt with by
showing people the consequences of their actions.
How can you reduce errors?
Error management should be tailored to suit specific
Staffing – is used to ensure that the right people are
contexts in particular organisations. The challenge for
placed in the right jobs. More importantly, it ensures that
organisations is to create environments in which people
people are recruited who can be trained in the skills and
can make their mistakes without dire consequences.
responsibilities to the level that will be required of them.
This means using the most cost-effective combination
of techniques across the five areas of human factors
Culture – is developed by the organisation, through
explored in this Guide, as follows.
its leadership (page 96), management (page 99)
and teamwork (page 103) so that people work in a
Design – is used either to ensure that people cannot
supportive, blame-free atmosphere. As a result, everyone
make certain sorts of mistake (eg by installing signal
develops a responsible approach to managing the
interlocking), or else to help users to review their
detection and correction of mistakes, reducing their
decisions before enacting them (eg a dialogue box that
consequences and preventing their re-occurrence.
asks ‘Are you sure you want to delete the selected file?’).
Automation may seem attractive because it designs
the human out. But automation is as much ‘fools gold’
as the elimination of error is. At best, it simply changes
Understanding Human Factors/June 08
Conditions – are considered with the aim of identifying
and reducing the mistake-making consequences of
motivation and morale (page 117), stress (page 120),
workload (page 125) and shift work (page 128). The
reduction is achieved mostly through effective design,
training and management (including self-management).
What error reduction techniques are available?
The majority of the techniques for human error reduction
require you to evaluate individual task performance
(actual or predicted) in terms of natural mental process
(eg a person sees something because they expect to see
it, overlooks something that is present, or focuses on one
area of the display to the exclusion of everything else).
The techniques also require you to evaluate the influence
of performance-shaping factors. These are sources of
influence on our behaviour such as fatigue and noise.
Several of the key techniques are listed below. You can
find out more about them in Part 3.
• Fault trees are used to depict system failures and
causes, and to estimate their probabilities.
• Human Error HAZOP (HAZard and OPerability) is
thorough and insightful – but very resource intensive.
• HEART (Human Error Assessment and Reduction
Technique) is well established in the railway industry, is
relatively simple to understand and use, and focuses on
factors that have the most influence on human error.
• Murphy diagrams are very similar to fault tree analysis
in that errors or failures are analysed in terms of their
apparent causes.
• SHERPA (Systematic Human Error Reduction and
Prediction Approach) is a human error prediction
Page 11
Part 2: Guidance
Human performance
technique that also enables tasks to be analysed and
potential solutions to errors to be presented in a
structured manner.
RSSB is currently developing a rail specific human
reliability assessment technique. This is initially driver
based, but will be extended to other operational groups
in due course.
Further information about human error
1 Edkins G.D. & Pollock C.M. (1997) The influence of
sustained attention on railway accidents. Accid. Anal. &
Prev. Vol 29, No. 4, pp553-539, Elsevier Science
2 HSE (1999) Reducing error and influencing behaviour
HS(G)48 (Second edition) HSE Books
3 Kirwan B. (1992a) Human error identification in
human reliability assessment. Part 1: Overview of
approaches. Applied Ergonomics Vol. 23(5), 299 – 318
4 Kirwan B. (1992b) Human error identification
in human reliability assessment. Part 2: detailed
comparison of techniques. Applied Ergonomics, 23,
371-381
5 Kennedy R. & Kirwan B. (1998) Development of a
Hazard and Operability-based method for identifying
safety management vulnerabilities in high risk systems,
Safety Science, Vol. 30, 249-274
6 OpsWeb, www.opsweb.co.uk/ (as of May 2008)
7 Pew R.W. Miller D.C. & Feehrer C.S. (1981) Evaluation
of Proposed Control Room Improvements Through
Analysis of Critical Operator Decisions. EPRI-1982.
Page 12
Electric Power Research Institute, Palo Alto, CA
8 Railway Safety Engineering Safety Management Yellow
Book 3: Application Note 3 Human Error: Causes,
Consequences and Mitigations Issue 1.0
9 Reason J. & Hobbs (2003) Managing maintenance
error, Ashgate Publishing
10 RSSB (2008) Good Practice Guide on Cognitive
and Individual Risk Factors, RS232, Rail Safety and
Standards Board
11 Vanderhaegen F. (2001) A non-probabilistic
prospective and retrospective human reliability analysis
method – application to railway system, Reliability
Engineering and System Safety 71 (2001) 1–13
12 Williams J.C. (1986) HEART – a proposed method for
assessing and reducing human error. In 9th Advances
in Reliability Technology Symposium, University of
Bradford
Why do people break rules?
Sometimes people break a rule because they don’t know
it exists, or they don’t understand it well enough, or they
fail to recognise that a situation demands it. Or perhaps
they simply forget that the rule exists. In all of these cases,
rule breaking falls into one or another of the categories
of human error discussed in the section on Why do
people make mistakes? on page 9.
Sometimes, however, people break a rule deliberately. This
means that the rule-breaking is not really an error, but a
violation. Why do people deliberately break rules?
Overwhelmingly, people do not break rules maliciously,
but for entirely rational reasons. In general, violations
result from the conflict between an organisation that is
attempting to control the behaviour of the workforce,
and the individual who is attempting to carry out their
task as easily as possible.
The UK railway industry classifies violations in the
following way:
• Routine violations
• Situational violations
• Exceptional violations
• Personally optimising violations
• Sabotage
Lawton’s (1998) research into Rule Book violations on
the UK railway system has revealed four main reasons
why staff break rules deliberately.
Understanding Human Factors/June 08
Part 2: Guidance
Human performance
methods of working than those who are fully
• Unusual circumstances. These can
‘Overwhelmingly, people rules within a system are often
accountable for the consequences of an accident or
impractical in everyday practice.
lead to exceptional violations
do not break rules
incident. Uncertainties in the allocation of authority,
and usually arise when a rare
•
Routine
short-cuts
.
Other
responsibility and accountability will increase the
combination of events gives rise
maliciously, but for entirely types of short-cut lead to routine
likelihood of violations.
to a novel response. An example
violations and occur when shortrational reasons.’
is the shunter who fails to show
cuts are regularly taken. An example Rule breaking usually has rather different outcomes
a hand danger signal to the
is
the
shunter
who
gets on and off the pilot engine
for the individuals that do it and the organisations they
driver to instruct them to remain stationary, before
while
it
is
moving.
Routine
violations
are
usually
high
work for. People tend to break rules deliberately when
walking between carriages. These are high-risk but
frequency
but
low
risk.
They
often
go
unnoticed
or
the benefits of doing so appear to outweigh the costs
low-frequency violations – the violations most often
(as long as they remain inside their own risk comfort
unremarked,
and
contribute
greatly
to
productivity.
associated with shunting accident fatalities. They occur
zone. See Why do people take risks? on page 15). For
People usually assume that the skill of the individual
when someone encounters a novel problem and
more than offsets any risk they might be taking. In such the organisation as a whole, however, rule-breaking on a
needs to use their knowledge to find a solution, eg
large scale may cause serious disruption to productivity
cases, individuals may also believe that the rules they
when a train needs to be coupled together on a bend.
and other losses due to the accidents that result overall.
are ignoring no longer apply to them. These routine
Take the example of car drivers who exceed the speed
transgressions end up being part of
• Situational short-cuts. Some types of short-cut can lead
limit. For society the economic and
the normal way of working within a ‘Uncertainty about
to situational violations and usually arise in difficult
social costs of road traffic accidents
particular work group.
conditions where someone sees an economical way
are huge. But from an individual’s
who is in charge
to keep the job going. An example is the driver who
perspective the personal costs appear
• Ineffective supervision. This can lead
does not stop even though they have lost sight of
unlikely and distant.
or
accountable
will
to various types of personally
the shunter during a movement. These violations are
optimising violation - or even
increase the likelihood
high-risk and high-frequency and are the result of the
sabotage in extreme cases (although
What can be done to reduce rule
employees’ immediate work environment. Violations
of
violations.
’
no
supervisory
system
–
however
breaking?
are inevitable in conditions where the work area
effective
–
can
prevent
a
determined
or equipment is poorly designed or under-staffed.
• Make sure the rule is necessary. Before trying to
saboteur). Ineffective supervision can lead to an
Such conditions make it difficult or impossible for
persuade people to follow a rule you should first see if
individual’s misjudgement, eg someone who breaks
staff to remain within the rules. While the job keeps
you can simplify the task or remove opportunities for
rules to prove to themselves, or others (probably
moving forward effectively, this type of violation is
error, and therefore the need for the rule.
mistakenly), that they have the additional skills
often ignored. Such violations may even be expected
needed to be in control of the risks. Alternatively, an
or endorsed by managers. It is another story when
• Make sure the rule is credible. Safety rules should be
environment with no recent accidents may be seen as
an accident happens and violators find themselves
about safety. If the main purpose of a safety rule is to
proof that the way people work is safe – which then
the subject of disciplinary investigations. Historically,
protect an organisation rather than the safety of the
produces complacency and a false sense of security.
industrial life in the UK has often revealed the
individual, its credibility will suffer (see Panel, Rules must
Finally, people who are not fully held accountable
operational difficulties created when staff ‘work to
be credible). It must be clear that the focus of the rule
for safety are more likely to adopt non-approved
rule’: such action typically highlights the way in which
is on safe behaviour, not compliance. Credibility will
Understanding Human Factors/June 08
Page 13
Part 2: Guidance
Human performance
Rules must be credible
In one accident in a process industry, a valve had to be
changed on a line carrying corrosive chemicals. The line
was emptied but a few drops of liquid remained. The
permit asked for goggles and gloves to be worn. The
fitter did not wear them and was splashed in the eye
by a drop of the chemical. At first sight this seems like
a violation, a deliberate failure to follow clear written
instructions. However, a look at the permit book
showed that every permit asked for goggles and gloves,
even for jobs on low pressure water lines in clean
areas. The maintenance crew therefore ignored the
instruction, deeming it to be more about management
covering itself rather than about their own protection.
Source: Kletz (unpublished), reproduced with permission
also suffer if it is out of date or out of sync with new
procedures – it needs to be reviewed at appropriate
intervals. Not doing so will increase the number of
routine violations.
• Make sure the rule is understood. People must be aware
of the rules themselves, how they fit with related rules,
the hazards that they are attempting to avert, and
the consequences of not observing them. This means
you need to pay serious attention to rule description,
training and dissemination.
• Make sure the rule is practicable. If the wrong method
is easier, or if the right method is impractical, people
will use the wrong method. You need to make it
possible for staff to plan their work to take the rule
into account, and provide the equipment necessary to
perform work according to instructions. You also need
to ensure that targets are achievable without short
cuts.
Page 14
• Make sure the rule is consistent with both organisational
and team goals. When the goals of a work group
conflict with the goals of the organisation this may give
rise to informal ways of doing things that encourage
infringements of the rules. The supervisor, being close
to the work group, may share the norms of the group
and therefore support such infringements. The aim of
the organisation should be to foster informal norms
that do not go against its goals.
• Make sure the rule is rehearsed right after training. Failing
to practice new rules soon after training – either
operationally or via simulation – simply wastes the
training resource. Either the rule itself will be forgotten,
or its perceived importance will be reduced.
• Make sure the rule is enforced. Rules must be supported
by effective monitoring of the work practices and
enforcement. You need to apply sanctions consistently
and fairly when non-compliance occurs. Increasing the
costs of violating will increase compliance.
Tools and techniques to reduce rule-breaking
A rule compliance toolkit has been developed by RSSB.
This toolkit helps railway managers identify the sources
of rule compliance problems – including violations – and
their solutions. The toolkit is available from the RSSB
website at www.rssb.co.uk (as of May 2008)
practices for reducing the potential for violations.
The methodologies were developed for the offshore
industries, but may be useful in other industrial sectors.
Further information about rule-breaking
1 Health, Safety and Engineering Consultants Limited
(2000) Techniques for addressing rule violations in
the offshore industries. Offshore Technology Report,
2000/096
2 Health and Safety Executive (1995) Improving
Compliance with Safety Procedures. HMSO, London
3 Lawton R. (1998) Not working to rule: understanding
procedural violations at work, Safety Science Vol. 28,
No.2, pp.77–95, 1998
4 Mason S. (1997) Procedural violations - causes, costs
and cures. In Human Factors in Safety— Critical
Systems, eds. Redmill, F. and Rajan, J. Butterworth
Heinemann, London
5 Mills A.J. & Murgatroyd S.J. (1991) Organizational
Rules: A Framework for Understanding Organizational
Action. Open University Press, Milton Keynes
6 Robens Lord (1972) Safety and Health at Work.
Report of the committee, 1970–1972. HMSO, London
The first document listed below in Further information,
commissioned by HSE, reports on the development
of a package of easy-to-use, but comprehensive,
methodologies that enable the non-specialist to identify
the underlying reasons behind violations. The project also
aims, as far as possible, to identify the best management
Understanding Human Factors/June 08
Part 2: Guidance
Human performance
Why do people take risks?
‘Nothing will ever be
attempted, if all possible
objections must be first
overcome.’ Samuel Johnson, essayist
Everything we do is a risk, since
there is always uncertainty
involved – however small. We
may be uncertain about the
information we are using to make
our decisions, or about whether
our chosen action will lead to the goal we wish to
achieve and what the cost will be. Fortunately, in most
aspects of our lives, we can use feedback about our
actions – especially our errors – to give us a better idea
of the risks involved in future actions. In any event, we
don’t want to avoid all risk, for risk adds excitement to
what we do. It removes boredom and adds perspective
to our decision making. It makes us conscious of our own
learning, growth and capabilities.
What affects risk taking?
closer to the car in front than ever
before (see Panel, Do organisations
compensate for risk?).
Some people appear to seek higher
levels of risk than others. They are
particular personality types, who are
also characterised by:
• reduced levels of self-control
• reduced long-term planning ability
• sensation seeking behaviour
• higher self-esteem
• high activity levels
• preference for personal freedom over adaptation of
social norms
People differ in the levels of risk they are prepared to
tolerate. Most of us actively seek to maintain our activities • need for independence
List reproduced with permission of authors and publisher from: Keinan
within a ‘risk comfort zone’. If things get too boring
G. Meir E. & Gome-Nemirovsky T. (1984) Measurements of risk takers’
or too risky, we modify our activities to compensate.
personality. Psychological Reports 55:163-167 © Psychological Reports
A good example of this is the motorist who becomes
1984
more familiar with a particular route and starts to drive
it more quickly. The risk arising from their uncertainty
Whether we are risk seekers or risk averse, a key
about the route is released for use elsewhere – in this
problem for all of us is that our perception of risk and the
case in the form of speed – so maintaining their exposure actual level of risk is usually quite different. This gap is due
to risk. Risk compensation theory helps
to any number of a wide range of factors
to explain why safety measures such as
that seriously distort our estimation of
‘Everything is
antilock braking systems, air bags, seat
how risky something is (see Panel on
sweetened by risk’ page 16, Factors affecting risk perception in
belt laws and speed regulations have
Alexander Smith, Scottish poet the railways).
not resulted in an overall reduction in
accidents. Instead, people use the reduced
risk these safety enhancements bring to drive faster and
Understanding Human Factors/June 08
Do organisations compensate for risk?
It seems that individuals operate their own ‘risk
economy’ in which they conduct their activities within
a self-defined risk comfort zone (see What affects
risk-taking?, this page). Does something similar operate
for whole industries? The aviation industry has suffered
what appears to be the same accident rate since the
early 1970s. This is allowing for increased flights, and
in spite of major advances in technology (eg Ground
Proximity Warning System and Airborne Collision
Avoidance System) and in training (eg Line Oriented
Flight Training and Crew Resource Management) as
well as reliability improvements in manufacture and
maintenance. As safety professionals discover more
about the complexities of accident causation and safety
management, it may be that they need to guard against
a new risk: that the organisational focus is on a ‘same
for more’ risk economy (ie same accident rate for more
productivity) rather than a concerted effort to reduce
the accident rate in absolute terms.
Further information in Keinen et al (1984)
How can you reduce risky behaviour?
Reducing risky behaviour is quite difficult. This is partly
because people conduct their activities in a way that
makes them feel safe – even if they are not. But it is
also because of the large number of factors that affect
people’s perception of risk. Together, these factors mean
that it is rarely effective to simply explain the real risk to
people. Instead, if people are not naturally sensitive to
risks (sometimes called being risk averse), we need to use
our knowledge of the factors to find effective ways to
persuade them.
Page 15
Part 2: Guidance
Human performance
Factor
Effect
Age
Both younger and older people often underestimate risk. They also overestimate their ability to deal with it.
Gender
Men tend to take more risks than women. They also perceive what they do as less likely to result in an accident.
Three lines of approach to reducing risky behaviour
involve selection, targeted communications and behaviour
modification programmes. You can find out more about
these in the paragraphs below.
Experience
More experienced people develop an increased awareness of risk and reduces the likelihood they will be involved in an
accident. However, habitual response to ‘routine’ hazards can reduce risk awareness.
Selection
Reward system
Payment by piecework may be good for productivity, but also increases risk-taking and accidents.
Overconfidence
Humans are naturally overconfident and unrealistically optimistic – probably a basic survival mechanism – which often leads us
to underestimate risk. A majority of us, for example, consider ourselves less likely than average to get cancer, get sacked from
our job, or get mugged. Obviously, only 50 percent can be ‘less likely than average’ to do anything.
Controllability
If we think activities are under the control of others, we think they are they are relatively more risky. For example, releases of
toxic chemicals by industrial facilities are judged to be riskier than activities under our own control eg driving, walking on the line.
Availability
Facts that readily come to people’s minds are usually rated as more probable than those less easily recalled from memory.
Anchoring
People often base their judgements on recent information that may be quite unrelated (see Panel on page 17, Vlad the Impaler).
Familiarity
Risks from unfamiliar activities (eg a radiation leak) are judged to be greater than risks from familiar activities (eg your job).
Fairness
Risks from activities believed to be unfair (eg the unpopular siting of a community incinerator) are judged to be greater than
risks from activities judged to be fair (eg vaccinations).
Benefits
Risks from activities that have questionable or indirect benefits (eg nuclear power stations) are judged to be greater than risks
from activities that have clearer, more personal benefits (jobs, monetary benefits, car driving).
Catastrophic potential
Risks from activities that can cause a significant number of simultaneous deaths and injuries (eg a rail crash) are judged to be
greater than risks from activities that cause deaths and injuries scattered across time and space (eg car accidents).
Understanding
Poorly understood risks (such as bad health due to exposure to radiation) are judged to be greater than risks that are well
understood or self-explanatory (such as slipping on ice).
Cause and effect
proximity
Perception of risk is reduced by the failure of the risk to manifest, eg an individual may indulge in a particular risky behaviour, like
red zone working, many times without suffering any apparent consequences.
Conflict avoidance
People underestimate risks that question their earlier decisions, eg cigarette smokers underestimate the risks associated with
tobacco compared with non-smokers.
Dread
Risks from activities that evoke fear or anxiety (eg exposure to cancer-causing agents) are judged to be greater than risks from
activities that do not arouse such emotions (eg accidents at work).
Trust
People or organisations which lack trust or credibility (eg companies with poor environmental track records) are judged to be
greater than risks compared with those that are trustworthy (eg effective regulatory agencies).
Accident history
Risks from activities with a record of major accidents or frequent minor accidents (eg level crossing incidents) are judged to be
greater than risks from those with little or no such history (eg new rules).
Factors affecting risk perception in the railways
Page 16
It may be helpful to consider people’s propensity for risk
as part of an organisation’s selection process. In this way,
organisations can screen people who are less inclined to
take risks in the first place. For example, the Zuckerman
Sensation Seeking Scale does this by asking people about
their thrill-seeking behaviour and boredom levels. This
psychometric test identifies those people who favour
risky activities in real life and who often intentionally
expose themselves to danger. Research at Loughborough
University showed that high scorers on the test are also
more likely to gamble and expose themselves to new
sensations eg through volunteering for novel activities.
Sensation seeking and risk taking are closely related.
Research among car drivers has shown that the Sensation
Seeking Scale can identify people who will commit
more moving traffic violations. Psychometric tests such
as this may form a useful decision aid – as part of a
comprehensive selection process – in making sure the
right people are available for safety-critical jobs.
Targeted communications
Communications designed to alter people’s attitude to
risk are likely to be ineffective unless they are guided by
the factors known to influence our perception of risk
(see Panel, Factors affecting risk perception in the railways).
O’Neill (2004) says that people fall into different groups
according to their sensitivity to risk, as follows.
Understanding Human Factors/June 08
Part 2: Guidance
Human performance
Vlad the Impaler
When did Vlad the Impaler die? Before asking the
person next to you, ask them to write down the last
three digits of their telephone number. Now ask them
for the date of Vlad’s demise. Chances are, there will be
a strong relationship between the two numbers, and
your colleague will have assumed that he lived in the
first 1,000 years AD. (In fact, he died in 1476, at the ripe
old age of 45). Why should this be? It turns out that if
people have little basis for judgement, they will tend
to use something from a recent activity – even though
it is completely unrelated. This is the phenomenon
of information anchoring, and it can enormously (and
unwittingly) alter a person’s perception of risk.
• Risk averse people are the most easily influenced by
communications about risk. They may be regarded as
natural risk managers who are ready to hear about
new risks and are likely to mobilise resources to deal
with them.
• Risk tolerant people tend to be ambivalent about
risk. Their response will depend on their personal
experience of the risk being addressed.
• Risk deniers are unlikely to take any form of action in
response to a communication. This may be because
they are in denial, or will not (or cannot) commit the
resources to deal with it – unless an incident occurs.
• Risk seekers are deeply convinced that are effective
risk managers. They see risks as a source of personal
challenge and adventure. Raising awareness of a risk
in such people may result in the probability of the risk
increasing rather than reducing.
Understanding Human Factors/June 08
Behaviour modification programmes
Safety behaviour is more influenced by the consequences
of such behaviour than by general attitudes to safety or
instructions to behave safely. The influence is stronger
if the consequences are positive rather than negative,
sooner rather than later, and certain rather than
uncertain. Behaviour modification programmes can
provide a framework for presenting these consequences
in a structured way. It does not really matter which
programme is used. However, to be effective, the
organisation’s safety culture must be capable of
Principles of behaviour modification
• Behaviour can be measured. To make measurement
possible, the behaviour you wish to change must be
carefully defined and observable.
• Behaviour is a function of its consequences. People
continue to behave as they do either because
the consequences reinforce this behaviour or
the consequences do not reinforce doing things
differently.
• Behaviour can be changed by providing appropriate
reinforcement and feedback. Positive reinforcement –
such as thanks, praise and support from colleagues
and management – promotes behaviour change,
whereas punishment in an organisational context –
such as blame, criticism and disciplinary action – is
almost always counter-productive.
• People whose behaviour you want to change need to
be involved in goal-setting. When people are involved
in setting challenging and achievable targets for
changing their behaviour, this adds to the positive
effects of reinforcement and feedback.
supporting the programme (we cover safety culture
in more detail in the section on Culture on page 91).
Evidence from experience with behaviour modification
techniques in a range of industries confirms that they
can lead to safer behaviour, and can result in reductions
in accident/injury rates. Behaviour modification
programmes have been most successful when they
have used a combination of all the principles of
behaviour modification (see Panel, Principles of behaviour
modification).
Successful behaviour programmes have not only
improved safety. They have also led to improved
productivity and business performance. The following
conclusions have emerged from the experience of
safety-critical organisations with behaviour modification
programmes:
• Behaviour modification programmes should not focus
exclusively on the behaviour of the individuals in the
workplace. They need to be sensitive to the demands
of the job and normal work practices.
• Much of the benefit comes from the engagement
process in identifying goals, setting targets, working to
achieve them, and measuring and providing feedback
on performance. The processes used to select and
implement a behavioural modification programme are
more important than the specific programme selected.
• It is also true that different behaviour modification
programmes may be suitable for organisations or
work-sites at different levels of maturity. It is important
to select a programme that is appropriate for the
maturity of the work-site.
Page 17
Part 2: Guidance
Human performance
Several companies in the UK Oil and Gas Industry
have used many of the available behaviour modification
programmes. These include Time Out For Safety (TOFS),
Advanced Safety Auditing (ASA), STOP and Care Plus see Fleming & Lardner (2000).
Further information about risk
1 Fleming M. & Lardner R. (2000) Behaviour
Modification Programmes: Establishing Best Practice,
The Keil Centre, Edinburgh. (Research performed for
the HSE)
2 Keinan G. Meir E. & Gome-Nemirovsky T. (1984)
Measurements of risk takers’ personality. Psychological
Reports 55:163-167 © Psychological Reports 1984
3 Loughborough University, Ergonomics Ltd & Aston
Business School, (2004) Recidivist risk takers who
work at height. HSE.
4 O’Neill. P. (2004) Developing a risk communication
model to encourage community safety from natural
hazards. Discussion Paper Jun 2004 for State
Emergency Services, Australia
5 Wilde G.S.J. (1998) Risk homeostasis theory: an
overview. Injury Prevention 4, 89-91 British Medical
Journal Publishing Group
Page 18
Why do accidents happen?
If you were employed by the railway industry in the
year 2002–03, there was around a 1 in 50 chance that
either you or one of your colleagues was involved in a
workplace accident requiring at least three days off work
to recover. Over the same period, you would have been
twice as safe working in the construction industry and
three times as safe in the agriculture industry (HSE).
Aren’t accidents due mostly to human error?
The number of workplace accidents on the railways
reflects the fact that it is a relatively dangerous place. But
why do these accidents happen? Obviously people make
errors – see Why do people make mistakes? on page 9.
And just as clearly, human errors are intimately involved
in accidents. In fact, they are so involved that it seems
reasonable to blame them for the accidents that happen.
However, human errors don’t provide anything like the
full story of how accidents happen.
Before 1947, investigations of military aviation accidents
had concluded that pilot errors were the cause of
crashes. But then two psychologists, Fitts and Jones,
looked more closely at what pilots did in the cockpit.
They realised that the design of the instruments and
controls was producing misreadings and actions that
had never been intended by the designers. The pilot
errors were not random events. Rather they resulted
from understandable, regular and predictable aspects
of the designs they were faced with. What is more, the
errors occurred much more often than accidents did.
Significantly, disasters and near misses usually occurred
only when these human errors occurred in combination
with other factors or other circumstances.
‘Rather than being the main
instigators of an accident, operators
tend to be the inheritors of system
defects …Their part is that of adding
the final garnish to a lethal brew
whose ingredients have already been
long in the cooking.’
James Reason, psychologist
One solution was re-design. This means that modern
approaches to user-centred design (page 25) are essential
to avoiding the kind of error discovered by Fitts & Jones.
But it is not enough. And neither are better training
(page 55) and selection (page 79) – vital as these are
to a comprehensive solution. While there is a place for
all of these approaches, they all centre on the user. The
breakthrough in understanding accidents came when
they were seen not just as something bad that happened
to users, but as products of the way in which the whole
system worked.
Not only was it recognised that accidents tend to be the
result of a complex chain of contributory events, but also
that some of the factors contributing to an accident are
permanently present in normal working conditions. So
the question became ‘What are all of these contributory
factors, and how do they all come together sometimes to
create an accident?’
To answer this question, James Reason came up with his
now widely-known ‘Swiss cheese’ model.
Understanding Human Factors/June 08
Part 2: Guidance
Human performance
Reason’s (1990) ‘Swiss cheese’ model of accident causation
Reason (1990) says that systems have multiple layers of
defence against hazards and errors. It is only when the
failures in these defences line up with each other that
an accident or incident results. The last line of defence
is a person’s ability to compensate for mistakes (page
9) (eg those arising from bad interface design (page 33)
or workplace design (page 41)) and adverse events (eg
those arising from increasing workload (page 125) or poor
working conditions, resulting in stress (page 120) and
even more errors).
The involvement of a human being is a positive benefit
to the normal functioning system as well as being a
source of increasing weakness as abnormal or degraded
conditions nibble away at their ability to cope. Skilled and
Understanding Human Factors/June 08
motivated operators of equipment can be very effective
last-line defences under
surprisingly adverse conditions
as they search to prevent a
bad situation getting worse.
To see this happen, you only
have to watch an experienced
signaller at a busy panel in the
rush hour having to cope with
a track circuit failure.
How can you identify the ‘holes in the cheese’?
Some of the ‘holes in the Swiss
cheese’ are active failures
of human or mechanical
performance, and others are
‘latent’ conditions, such as
management factors or poor
system design. However, it is
clear that if steps are taken
in each case to reduce the defensive gaps, the overall
chance of accidents occurring will be greatly reduced.
A key tool in accident investigation is root cause analysis.
There are a number of methods for this, including Fault
Trees and Why-Because analysis (see Part 3). An important
human factors tool in this area is HFACS (Human Factors
Analysis and Classification System). Based on Reason’s
approach, the HFACS framework provides investigators
with a comprehensive tool for identifying and classifying
the human causes of incidents. This tool is relatively easy
to use, though its designers have assumed that its users
will be human factors experts. (For this reason we give
the key reference (Shappell & Wiegmann, 1997) but not a
detailed summary in Part 3 of this Guide.)
Latent failures at the managerial level can be reduced
by organisational planning. Psychological failings can be
reduced by paying attention to the types of task that are
required of workers, eg through task analysis, (page 47).
Unsafe acts can
be reduced by
‘To err is human; to
good workplace,
blame it on the other
equipment and
interface design.
guy is even more
human.’ Bob Goddard, rocket
scientist
It’s never possible to identify all the ‘holes’ that may
develop in the future. But paying attention to all the areas
outlined in this Guide, together with the best engineering,
management and health and safety practice, will do a lot
to stop the holes all lining up catastrophically.
Looking back, it is possible to discover a great deal about
the size and origin of the holes – and how they came to
be lined up – through informed accident investigation.
The use of tools such as these in support of rail accident
investigations is being guided by the new Railways
(Accident Investigation and Reporting) Regulations 2005,
in which human factors considerations are required to
cover specifically the ‘man-machine-organisation interface’.
A guide to the human factors aspects of rail incident
investigation has been incorporated into formal SPAD
investigation guidance (see Further information). (For an
overview, see Panel, HF checklist for accident investigations).
Page 19
Part 2: Guidance
Human performance
HF checklist for accident investigations
Best practice in conducting interviews following an accident
Personal factors
Preparation
Eliciting and storing information
•Fitness and health (illness, disability, medication, drugs and alcohol etc)
Investigators must be prepared to deal with details of a
more private nature concerning the interviewee, if these
relate to the accident. Interview preparation is very
important, not only to avoid the need for re-interviewing
but also to obtain full co-operation and good-quality
human factors information without having to go into too
much detail at a private level. It helps to gather as much
information as possible before the interview, listen to
any voice recordings, look at other records and plan the
questions to ask. Explaining the interview structure to
the interviewee may also help them to understand the
aims and phases of the interview.
An effort should be made to ask open questions and
use leading (or closed) questions to get confirmation
of details. Open questions start with the words ‘what’,
‘who’, ‘when’, ‘why’ or ‘how’ (‘How was your stress level at
that time?’). In comparison, closed questions prejudge the
response, can usually be answered by yes or no, and start
with a verb (‘Didn’t you feel that you were quite stressed?’).
•Lifestyle (financial worries, births, marriages, deaths etc)
•Morale and motivation
•Competence (task demands, assessment, training, complacency etc)
Mental factors
•Perception
•Vigilance and attention
•Memory
•Decision-making process
•Situational awareness
Team factors
•Communication
•Leadership
•Teamwork
Working condition factors
•Management (supervision, rules, procedures, planning etc)
•Environmental aspects (noise, light, etc)
•Equipment (design, training, maintenance, malfunction etc)
•Fatigue (shift work schedule, overtime, breaks, etc)
•Workload (too much, too little etc)
In 2002, a SPAD Hazard Checklist was developed for
use across the railway industry (revised in 2004).This
identifies both working practices and communications
procedures that should be adopted by both signallers and
drivers to help prevent SPADs occurring. Over and above
these guides, it may be helpful to consider more general
human factors guidance on how to get the most from
accident investigations.
Page 20
Duration and atmosphere
Interviews ought not to be too long (two hours
maximum) and performed as soon as possible, as the
interviewee may forget the details of the incident. What
is more, the interviewee’s ‘rationalisation bias’ (the
tendency to find a plausible explanation or justification
of the facts) will increase if there is a long time between
the event and the interview. During the interview, it is
important to set a tone that creates confidence without
causing confusion about the role of the investigator. The
place should be comfortable, with style and vocabulary
adapted to the interviewee. However, it is important to
maintain a professional approach.
Note
This general guidance is complemented by a specific recommended
procedure for use by Driver Standards Managers in interviewing
drivers following a SPAD. (see Further Information).
Active listening is another key element in interviewing
technique. It implies paying attention both to what is
said and to body language, and consists of repeating or
re-wording the interviewee’s answer in order to avoid
misunderstanding and to obtain as much information as
possible. It is also worth considering the use of some
means to prompt recall – such as a track layout diagram,
transcripts of communications, or photos (or a video) of
the reconstruction.
Human factors checklists (see Panel, Checklist of human
factors issues for investigations) may also be used for
guiding the discussion. It is important not to jump to
conclusions and to search for elements that might
disprove the hypothetical explanation of what happened
rather than confirm it. It could be useful to keep track
of interviews by tape recording them or taking notes.
However, this may change the atmosphere and even
be counterproductive as interviewees may express
themselves less easily. When recording or note-taking, it
is important – if a climate of confidence is to be created
– to give the interviewee right of access to the recorded
material. National legislation may cover this point in the
case of a serious accident.
Understanding Human Factors/June 08
Part 2: Guidance
Human performance
Things to remember about causes
•The investigation must go far deeper than the human error that is
often the last thing to happen before an accident.
•There is a chain of elements/factors that precede the accident, and all
of them have contributed to it.
•No single contributor sufficiently explains the accident.
•Each contributor has by itself increased the probability of the accident.
•Some contributing factors are instantaneous (eg track circuit failure)
while others have existed for a long time (eg shortage of experienced
signallers).
•Many factors exist all the time and do not normally lead to an accident.
•Some permanent factors (eg the shortage of experienced signallers
or the high density of train movements) may become very critical in
degraded operations.
•Human error can take place at any hierarchical and organisational level
of the system – not just on the track, train, station or in the signal box.
•Understanding why the incident happened is much more instructive,
and allows a wider range of remedial actions. than focusing on the ‘final’
human error.
How do you get the most from accident investigations?
Post-incident interviews
The interview remains the most appropriate technique
for gathering behavioural and circumstantial data (see
Panel on page 20, Best practice in conducting interviews
following an accident).
Establishing causes
A key part of accident investigation is to use what we
know about the multiple contributing factors that are
generally responsible for accidents. It is not just about
human error. In fact, as the Guide says elsewhere, it is
better to think of errors as consequences, rather than as
causes (Why do people make mistakes? page 9). (Also see
Panel, Things to remember about causes.)
Understanding Human Factors/June 08
‘Being blessed with both
uninvolvement and hindsight it
is [tempting] to wonder how
these people could have been so
blind, stupid, arrogant, ignorant or
reckless.’ James Reason, psychologist
Investigator biases
Investigators need to remember that decisions and
actions that have a negative outcome will be judged more
harshly than if the same process had resulted in a neutral
or positive outcome. Unfortunately, we can expect
this result even when judges are warned about the
phenomenon and have been advised to guard against it.
Investigators must also be aware that it is a natural human
tendency to form hypotheses and then seek confirming
evidence. It is much more efficient to seek disconfirming
evidence. It has been famously said that you cannot prove
that there are no black swans by counting the many
thousands of white ones.
Investigators also tend to believe that people involved in
an incident knew more about their situation than they
actually did (see Reason’s tongue-in-cheek quote on this
page). They will tend to think that people should have
seen how their actions would lead up to the outcome
failure.
People’s behaviour should be assumed to be rational
(though possibly mistaken) from the point of view of their
knowledge and mindset and the multiple goals they were
trying to balance at the time. If we can understand how
these factors guided people’s behaviour, we can see how
they were likely to make an error given the demands of
the situation they faced.
Investigating consequences
Very often, you can learn much of value from an accident,
not just by establishing its causes but also by investigating
its consequences. For example, root cause analysis can
just as easily be applied to each effect of the accident as
well as the accident itself. Doing so will aid prevention by
focusing on the wider issues of workplace organisation
and regulatory failures.
Taking care of incidents
A near-miss is a situation where an error occurred but
was recovered before it developed into an incident
or accident. A near-miss is therefore an opportunity
to improve safety practice based on a condition,
or an incident with the potential for more serious
consequences (see Panel on page 22, How to make the
most of near-miss data). In the UK railway industry, CIRAS
(Confidential Incident Reporting and Analysis System)
has been set up to log reports of unsafe situations.
Confidential reporting systems can be an essential source
of information for aspiring High Reliability Organisations,
but it is not used as much as it might be (see Panel on
page 22, All’s well that ends well?).
Near-miss reporting is only really successful in an
organisation that wishes to develop a problem-solving
rather than penalising culture. Senior management needs
to be committed on a continuous basis to the issue and
make a point of publishing success stories. In addition,
whoever is responsible for the problem needs to be
involved in analysing it, for they will frequently have the
Page 21
Part 2: Guidance
Human performance
How to make the most of near-miss data
1 All reported near misses should be assessed for
seriousness. This is best done under the leadership
of a project manager who reports to a steering
committee responsible to the top management.
2 Why-Because Analysis is useful to capture the root
causes of the incidents and to identify effective
countermeasures.
3 The analysis is best performed by a focus group,
and should include the person responsible for
the problem as well as other stakeholders, eg
subcontractors
4 Each countermeasure’s implementation status
should be available to users and senior management.
Based on research by Braband & Brehmke (2002)
best ideas on how to prevent it recurring. This is only
effective in a positive ‘no blame culture’. The final thing
to get right is identifying the right countermeasure to
implement, through the right analysis. Why-Because
Analysis (a form of root cause analysis) is useful here.
All’s well that ends well?
Confidential reporting schemes are essential for
safety-critical organisations. But how well do they
work? A study of a chemical processing plant revealed
why safety-critical workers do not report recovery
from self-made errors to the near-miss reporting
system. The main reasons were that they made and
then totally recovered from the mistake themselves,
that they didn’t think the near-miss system was for
reporting their particular errors; and that there was no
bad consequence to their error. Surprisingly, failures
to report these errors were shown not to be due
to staff worrying about being blamed or shamed
– they just didn’t think reporting it to be worthwhile
or relevant. These results represent a fundamental
problem for organisations: how to communicate their
interest in successful operator recoveries and so move
away from an ‘all’s well that ends well’ philosophy that
helps no one. The secret probably lies in management
demonstrating genuine pride and value in the expertise
of their operatives through appreciation and continuous
training schemes.
Based on research by Van der Schaaf & Kanse (2002)
Techniques for accident/incident investigation
HFACS (Human Factors Analysis and Classification
System) provides investigators with a comprehensive
tool for identifying and classifying the human causes of
incidents – but you need to be a human factors expert to
use it (see Shappell & Wiegmann, 1997). HFACS helps to
determine whether the individual, the line supervisor and/
or management had the responsibility for preventing the
error. It has been applied to the analysis of human factors
data from approximately 1,000 military aviation accidents.
Fault trees. A form of root cause analysis – but be careful
Page 22
that the technique does not lead you to stop the analysis
at the inevitable human error(s) that immediately
preceded the accident/incident.
Why-Because Analysis. This is an easy-to-use form of root
cause analysis recently simplified for use by line managers
in the railway industry. It has been applied to the analysis
of near-misses in the railway industry in order to identify
effective countermeasures.
‘The point of learning about human
error is not to find out where
people went wrong; it is to find out
why their assessments and actions
made sense to them at the time.’
Sidney Dekker, Professor of Human Factors &
Aviation Safety
Further information on accident investigation
1 HSE Enforcement Statement/Quality Statement for
Continuing Aim 2: Document G - Major Incident
Response And Investigation Policy And Procedures,
April 2001
2 Human Engineering Ltd, (2004) User Guide for
Human Factors SPAD Hazard Checklist: Issue 2 HEL/
RSSB/041123/RTB02, RSSB
3 Park, L. (2004) Work-related deaths - Investigators
Guide, HSE. Produced to assist those tasked with
investigating deaths in the workplace
4 The Railways (Accident Investigation and Reporting)
Regulations 2005, Statutory Instrument 2005 No. 1992,
www.opsi.gov.uk/si/si2005/20051992.htm#6 (as of May
2008)
5 Reason J. (1990) Human Error. New York: Cambridge
University Press
Understanding Human Factors/June 08
understanding
© Four by Three 2003
Design
human factors
understanding
Design
human factors
© Four by Three 2003
Part 2: Guidance
Design
Design
‘Every system is
perfectly designed to
achieve the results it
gets.’ Don Berwick, MD
Design is by far the cheapest and most
effective way for a system or organisation
to benefit from paying attention to human
factors. If a system delivers exactly the results
required by an organisation, it represents a
happy convergence of user requirements,
designers’ intentions and practical implementation.
Focus on design
Often, a system’s results are less
than satisfactory. However, no
matter what they are, they will
always flow directly from the way
the system is configured – whether
this configuration is intentional or
not.
There are two implications. First, if you want different
results, you need a different design for it. Second, since a
system’s results are produced by the joint behaviour
of equipment and its users, designers need to
successfully account for both.
This diagram focuses on design. It shows
six topics (in the middle red ring), and
identifies the main human factors
questions that this part of the
Guide answers (in the outer grey
ring). At the end of every section,
you will find a list of sources
of further information that will
provide more detail. In addition,
Part 3 gives further detail on key
human factors techniques that
are mentioned throughout this
Guide.
User-centred design
Why is user-centred design important?
Both individual users and the organisation as a whole will
perform better if users are involved in the design of their
equipment, tools and working environments. Involving
users in evaluating the design product at an early stage in
the design process will help to ensure that the product
is best suited to its purpose. It will also minimise the
time, effort and costs associated with making subsequent
design changes. Once a system is in development,
correcting a problem can cost an estimated ten times
more than fixing it during design; once a system is being
used, it can cost 100 times more.
What is required to do user-centred design?
The users are the people (eg drivers) who use the
product (eg a cab design), whether frequently or just
occasionally. They may also be the people who are
affected by the use of the product (eg passengers) or
make decisions about its purpose (eg TOC managers).
They don’t all have to be represented on a design team,
but the team must take all their views into account if
a product (ie anything that is designed) is to be usercentred.
For design to be both user-centred and sound, the design
team must ensure that:
• the end users (those who will actually use the
product) take part in the design process
• data on the needs of all types of user – ie all
stakeholders – is collected and analysed (see Panel on
page 26, What is a stakeholder analysis?)
Understanding Human Factors/June 08
Page 25
Part 2: Guidance
Design
What is a stakeholder analysis?
A stakeholder is a person or organisation with an
interest or concern in something – such as a to-bedesigned system (the project). Stakeholder analysis is a
tool for understanding a project’s potential impact by
identifying the stakeholders and assessing their interest
in the project. Keys steps of a stakeholder analysis are:
1 Establish the objectives of the project
2 Draw up a stakeholder table that identifies
stakeholders and their interests (both overt and
hidden). Each stakeholder should be asked:
What are your expectations of the project?
What benefits are there likely to be for you?
What resources do you wish to commit (or avoid committing) to
the project?
What other interests do you have that may conflict with the
project?
How do you regard other stakeholders?
3 Assess the likely impact of the project on each of
the interests (plus, minus or unknown). A workshop
is a good way of doing this. At the workshop, it is
often helpful to use something like cognitive mapping
to capture how the stakeholders cluster by role and
interest, plus the degree and direction of influence
they have on each other regarding the project.
It’s worth remembering that the likelihood of a
stakeholder being noticed and involved will be down
to their relative power to influence the project, the
urgency with which they regard the project, and the
relative legitimacy of their interest in the project.
Approaches for involving stakeholders in design and development
Approach
Resulting information
When to use
Background interviews,
questionnaires and checklists
Information about the purposes and constraints of the target users; prototype
evaluation. Questionnaires should not be used for primary data capture, but to
confirm requirements – also useful as a checklist to help structure interviews.
Beginning through mid-point in the
design cycle
Analysis of organisational
and business requirements
Organisational policy that is understood and transferred to the design project. A
key benefit is translating from the abstract to the specific, eg ‘make a test system for
signalling equipment’ becomes ‘support on-site testing by maintenance staff ’.
At the beginning of the design
project
Task sequence interviews
(eg heuristic analysis) and
questionnaires
Information about the target sequence of activities within the system’s operational
contexts and conditions.
Early in the design cycle
Observational analysis
Information about the physical environment in which the product will be used, eg
Japanese auto engineers spent significant time in the homes of typical customers to
understand user needs and values.
Early in the design cycle
Analysis of user problem
reports for existing systems Information about the negative aspects of related existing systems through problem
reports and user suggestions – most useful if the new product is an evolution
rather than an innovation. Statements like ‘the display is too bright at night’ or ‘the
rules conflict with each other’ are very revealing.
Early in the design cycle
Analysis of existing systems
Information about the existing elements of similar systems to the one to be
developed. Investigate different industries if necessary – don’t be afraid to ask for
access. Ask users what they like and dislike.
Early in the design cycle
Previous user modifications
Information about the way in which users have developed ad hoc fixes through
engineered solutions to original equipment. These modifications will often give
strong clues about previous design shortcomings and better understanding of user
needs. They can often be reverse engineered into new design requirements.
Early in the design cycle
Focus groups and
workshops (eg using
brainstorming, hexagons,
cognitive mapping)
Information about requirements and prototype feedback from representative
stakeholders, including operators, supervisors, managers, trainers, engineers and
maintainers.
Early through mid-point in the
design cycle
Role playing, walk-throughs,
and simulations
Information about the effectiveness of different designs and further insight into user
needs and expectations – particularly useful for brand new, rather than evolved,
systems.
Early through mid-point in the
design cycle
User group meetings
Information about the use and operation of existing systems - most useful if the
new product is an evolution rather than an innovation.
All through design cycle
Usability testing
Information about the usability of the system
At the end of each prototyping
stage and as the final stage of the
design cycle
Sources: Stevens et al (1998) reproduced with permission; also summarised from research by Preece et al (2002)
4 Define options for managing the interests.
Adapted from: Department for International Development (1995)
© Crown Copyright, reproduced with permission
Page 26
Understanding Human Factors/June 08
Part 2: Guidance
Design
• data on the tasks that users (both operators and
maintainers) perform in using the product is collected
and analysed
• the wider technical and economic system to which the
product belongs is considered
• the process involves user feedback and re-design
where necessary
• the best possible outcome for the greatest number of
people is achieved within reasonably practical limits.
How is user-centred design done?
A ‘user-centred design’ approach requires that the design
of equipment and systems is based on understanding
the needs and characteristics of its users. The design
process needs to involve stakeholders in a continuing
process of consultation and testing, which draws on all
the available data on the purposes, needs, capabilities and
The systems engineering process
From Stevens et al, 1998 reproduced with permission
Understanding Human Factors/June 08
limitations of humans. The Panel on page 26, Approaches
An important part of defining user requirements is to
for involving stakeholders in design and development, sets
remember that there is always more than one type of
out techniques for involving users in the
user with a stake in the system to be
‘We
often
don’t
design and development of a product at
designed. The end users are the people
various stages in the design cycle.
notice good usability, who actually use the system, and
their operational, physical and mental
but
we
almost
always
capabilities must be accommodated.
How does user-centred design fit
notice poor usability.’ However, the end users and the
with systems engineering?
new product must together deliver a
Many designers will be familiar with
solution that is usable by other parts of the organisation
the systems engineering process, which is a structured
(or perhaps by the organisation’s customers). It is a
way of creating effective solutions and managing their
good rule of thumb that designers fully understand the
technical complexity. The diagram on this page illustrates
requirements of both the product’s end users and the
the main stages of the process. The earlier stages of the
end users’ ‘customers’ if their design is to be effective.
systems engineering process are concerned with defining
In fact, it is very important for designers to understand
the requirements and the product to be built. Later, the
the interests of all the stakeholders in the system. This
emphasis switches to integration (where the product
underlines the importance of carrying out a proper
is assembled) and verification (where the assembled
stakeholder analysis (see Panel on page 26, What is a
product is tested) before delivery to the users. The blue
stakeholder analysis?)
boxes show where hard science-based research (eg
physics, engineering, chemistry, materials) development,
What is a user requirement?
design and manufacturing must take place. The amberDesigners need to distinguish between three different
coloured boxes show where
types of user requirement. The first type is to do with
human factors considerations
user aims. Designers must understand what users need
need to be plugged in – via the
the product for, and how it needs to fit with their general
techniques outlined in the table.
workflow and that of the organisation as a whole.
To ensure the most effective use
The second is to do with user characteristics. This means
of these techniques, a Human
that designers must understand what the capabilities,
Factors Integration Plan (HFIP)
limitations and expectations of the users are, including:
needs to be constructed (see
Panel on page 28, What’s the
• visual capabilities
plan?).
• degree of task and system-specific expertise
Page 27
Part 2: Guidance
Design
What’s the plan?
Below are the main elements of a Human Factors Integration Plan
(HFIP). If you operate your design project along these lines, you will go a
long way towards effectively integrating human factors into your systems
engineering process.
1 Responsibility – Who is in charge of HF for the project?
2 Stakeholders – How are they to be identified and consulted?
3 Contractors and sub-contractors – How will they ensure
sufficient attention to HF? (What and where is their HFIP?)
4 Coordination – How are HF aspects to be coordinated across
all parties involved? How will decisions made be followed up and HF
issues closed off?
5 Management – How will HF considerations be incorporated into
the regulatory consultation and approvals processes?
6 Operational concept – When will it be required and what will
it contain?
7 Legacy information – What can be utilised from existing/similar
systems to identify key aspects of users, processes, equipment,
working conditions, and safety?
8 Design options assessment – How will it address user
requirements, including capabilities, limitations, reliability, workload,
health and safety, and hazard prevention?
9 Standards – What HF standards and principles will be used during
detailed design?
10Operability trials – What criteria will be used and how will
performance be measured? What is the end user trials schedule?
How will trials feedback be captured and used?
11Support – How will documentation, help, training development and
delivery be implemented and when?
12Evaluation – How will data on the performance of the system in
use be collected, analysed and used?
Adapted from London Underground Ltd (2002) with permission
• expected maintenance schedules and required levels
of maintainer expertise.
The third type of user requirement is to do with user
values. If a new piece of technology is to be successfully
adopted, it must take account of what motivates users
and what can just as easily turn them off:
• usefulness – how far users perceive the innovation as
being better than its predecessor
• compatibility – how far users perceive the innovation as
being consistent with their existing values, needs and
past experiences
• ease of use – how far users perceive the innovation as
being difficult to use
• user stereotypes and expectations of what the
product is for and how it works
Page 28
• Absolute reference – what is the unique code that
identifies the requirement?
• Source – who asked for the requirement?
• Ownership – who needs this requirement?
• Priority – how important is the requirement?
• Performance – how well must this requirement be
met?
• Urgency – how soon is the requirement needed?
• self-image – the extent to which a person’s image is
enhanced by the innovation ie how far it fits with (and
promotes) a person’s self-concept in the eyes of their
peers (its ‘coolness’)
• Stability – is the requirement clear and accepted
enough for design work to start?
• trialability – the degree to which users can try working
with the innovation (eg by a trial period) before having
to make a total commitment to adoption
• Acceptance criteria – what test(s) will satisfy the user
that the requirement is met?
• clarity – the degree to which the results of an
innovation are clear and communicable to others.
(See Panel, Checklist for user requirements analysis.)
• extent and regularity of system use
• experience with similar equipment
Checklist for user requirements analysis
You may find it helpful to use the following questions in
documenting each user requirement.
‘It is vital to collect
feedback from users as an
ongoing process.’
• Verifiability – how will the final product be tested
against this requirement?
Source: Stevens et al (1998) reproduced with permission
How can usability be tested?
A key technique of user-centred design is assessing the
designed product’s usability – ie the extent to which
it allows the user to reach their task objectives. Good
usability is a conscious and deliberate design goal. We
often don’t notice good usability but we almost always
notice poor usability. The idea of usability is highly relevant
to the design of human-machine interface, hand tools and
equipment, as well as the design of workplaces.
Understanding Human Factors/June 08
Part 2: Guidance
Design
In assessing the usability of any product, we need to
consider who the specified users will be, what the
specified goals will be and what environment(s) the users
will be using the product in.
The usability assessment process
The assessment process should:
• involve real users - both operators and maintainers
(see Panel, Maintainers are users too!)
• give the users real tasks to carry out – including
normal, degraded, abnormal and emergency conditions
• enable testers to observe and record user actions
• enable testers to analyse the data that is collected and
make changes in response.
The three key criteria for assessing usability are:
1 Effectiveness – how well does the product do the job?
Does it enable users to: complete the task?; do the
task well?
2 Efficiency – how easy and/or fast can users get
things done with the product? Does it enable users
to: complete the task quickly?; complete the task
reasonably easily?
3 Satisfaction – are users satisfied by the way in which
the product supported their work? Do they like it?
In practice it is very difficult for designers to know or
imagine all the usability criteria that are important to
users. This is why it is so important to collect feedback
from users as part of an ongoing process to improve the
Understanding Human Factors/June 08
Maintainers are users too!
Maintainers are a key group of stakeholders in any
new or updated system. Attention to design from
the maintenance perspective will involve many
interdependent decisions about modularity, accessibility,
maintenance working conditions and inspection
schedules, maintainer knowledge and skill demands,
training requirements and staff availability, as well as
safety. Research has produced good human factors
guidelines for the kinds of problem that afflict the
maintenance environment. As well as providing a handy
checklist for maintenance workers and their managers,
they are also a useful resource for designers. Anything
that can be done at the design stage to avert the more
common maintainer problems will substantially increase
the usability and effectiveness of the system (see Panel
on page 30, The Dirty Dozen).
typical task environment so that the following data can
be collected:
- time for users to learn a specific function
- speed of task performance
- type and rate of errors by users
- user retention of commands over time
- subjective user satisfaction.
• walk-throughs – in which a group of users step
through tasks, and problems are noted for discussion
• focus groups – to discuss aspects of the product both
before and after it is in use.
Recommended usability testing techniques
• System Usability Scale (SUS)
• Software Usability Measurement Inventory (SUMI)
operability and maintainability of the product.
• User trials
Techniques for usability testing
• Focus groups
Among many techniques for usability testing are:
• Workshops
• think-aloud techniques – in which the user is asked to
describe all the steps they take in carrying out a task
• Interface surveys
• videotaping – so that designers can review what users
do, and see where the problems are in their designs
• Questionnaire for User Interface Satisfaction (QUIS)
• Cognitive Walk-through.
• interviews and usability questionnaires – enabling
designers to evaluate what users like and dislike about
the design and increase their understanding of any
problems
• testing and data logging – where the tests require
typical users to perform typical standardised tasks in a
Page 29
Part 2: Guidance
Design
The Dirty Dozen: Common maintainer human factors problems – and solutions
The Dirty Dozen: Common maintainer human factors problems – and solutions
Problem
Solution
Problem
Solution
Lack of
communication
•Use logbooks, worksheets, etc, to communicate and remove doubt
Stress
•Be aware of how stress can affect your work
Complacency
Lack of knowledge
•Discuss work to be done or what has been completed
•Stop and look rationally at the problem
•Never assume anything
•Determine a rational course of action and follow it
•Train yourself to expect to find a fault
•Take time off or at least have a short break
•Never sign for anything you didn’t do [or see done]
•Discuss your problem with someone
•Get the relevant training
•Ask fellow workers to monitor your work
•Exercise your body
•Use up-to-date manuals
•Ask a technical representative or someone who knows
Distraction
•Remain focused on finishing the job
Lack of
awareness
•Double inspect by another or self
•When you return to the job, always go back three steps
•Use a detailed check sheet
Lack of
teamwork
•Discuss what, who and how a job is to be done
Fatigue
•Be aware of the symptoms and look for them in yourself and others
•Be sure that everyone understands and agrees
•Plan to avoid complex tasks when you will be tired eg at end of shift
•Sleep and exercise regularly
•Ask others to check your work
Lack of parts
•Check suspect areas at the start of the inspection and ensure parts availability
•Order and stock anticipated parts before they are required
•Know all available sources of parts and arrange for pooling or loaning
•Maintain a standard and if in doubt do not sign off the job
Pressure
•Be sure the pressure isn’t self-induced
•Communicate your concerns
•Ask for extra help
•Just say ‘No’
Lack of
assertiveness
Page 30
•Only sign for what is serviceable
•Refuse to compromise your standards
•Check to see if your work will conflict with an existing modification or repair
•Ask others if they can see any problem with the work done
•Mark the uncompleted work
•Lockwire where possible or use torqueseal
•Think of what may occur in the event of an accident
Norms
•Always work as per the instructions or have the instruction changed
•Be aware the ‘norms’ don’t make it right
Adapted from: Dupont (1997), reproduced with permission
Further information about user-centred design
1 BS 6548-2:1992, IEC 60706-2:1990. Maintainability of
equipment. Guide to maintainability studies during the
design phase
2 BS EN ISO 13407:1999. Human-centred design
processes for interactive systems
3 BS ISO/IEC 15288:2002. Systems engineering
– System life cycle processes.
4 CAA (2002) An Introduction to Aircraft Maintenance
Engineering Human Factors for JAR 66, CAP 715,
www.caa.co.uk (as of May 2008)
5 Dumas S.J. & Redish J.C. (1993) A practical guide
to usability testing, Norwood: Ablex Publishing
Corporation
Understanding Human Factors/June 08
Part 2: Guidance
Design
6 Dupont, G. (1997)The Dirty Dozen Errors in
Maintenance. In: proceedings of the 11th Symposium
on Human Factors in Aviation Maintenance
7 As of May 2008, www.usabilityhome.com/ Summarises
several usability evaluation methods organised under
the three types of: Testing, Inspection, and Inquiry
8 As of May 2008, http://jthom.best.vwh.net/usability/
usable.htm is a very useful ‘how-to’ website for
usability testing
9 ISO/TR 16982:2002. Ergonomics of human-system
interaction – usability methods supporting humancentred design
10 London Underground Ltd (2002) Manual of Good
Practice in Human Factors Integration, Rev A1, M
1035 R2.
11 Nielsen J. (1993) Usability Engineering. Morgan
Kaufmann
12 Norman D. (1988) The Psychology of Everyday Things.
Doubleday, New York
13 Preece J. Rogers Y. & Sharp H. (2002) Interaction
Design: Beyond human-computer interaction. John
Wiley, New York, NY
14 Stevens R. Brook P. Jackson K. & Arnold S. (1998)
Systems engineering: coping with complexity, Prentice
Hall, Europe
Understanding Human Factors/June 08
Equipment design
What makes good equipment?
Essentially, a good piece of equipment is one that is fitted
to its purpose. This usually means that it is easy for people
to operate and maintain. No doubt you frequently come
across, and suffer from, numerous examples of poor
equipment design in your everyday life, ranging from
high-tech items such as video recorders and computers
to basic items such as taps. The design of equipment for
the workplace is often no better. In fact, poor design
– particularly poor control room equipment design – has
contributed to many major air traffic, marine, military and
industrial accidents.
Much has been known for a long time about what
constitutes good and bad equipment design. But there
continue to be many examples where designers do
not apply good human factors principles. One dramatic
example was the accident at the Three Mile Island nuclear
power plant in 1979 in which the nuclear core came
close to a catastrophic meltdown. The accident resulted in
permanent closure of the facility and sweeping regulatory
and operational changes involving emergency response
planning, reactor operator training, human factors
engineering, radiation protection, and many other areas of
nuclear power plant management. The main cause of the
accident was that the control room was poorly designed,
with problems that included:
• controls positioned too far from the instrument
displays that showed the system’s condition
• cumbersome and inconsistent instruments that often
looked identical and were placed side-by-side, even
though they controlled very different functions
• instrument readings that were difficult to read,
obscured by glare or poor lighting or actually hidden
from the operators
• inconsistencies in the meaning of lights and the
operation of levers or knobs (ie pushing a lever up
may have closed one valve, while pulling another lever
down may have closed another one).
What are the principles for equipment design?
The fundamental principles of good equipment interface
design are:
• Visibility – placing the controls where the user can
easily see them, with adequate lighting for doing so
• Feedback – telling the user when their input has
resulted in the system doing something, as when route
lights illuminate on an NX panel in response to the
signaller setting a route – but see Panel, Too much user
feedback!
Too much user feedback!
Designers of the railway 375 stock made the AWS
and CSR audible levels loud enough to cater for the
worst case scenario to ensure they attracted attention.
But drivers reported that these were excessively loud,
resulting in earache and headache by the end of a shift.
Design modifications had to be set in motion.
• Natural mapping – helping the user to understand how
something works by mapping the relationship between
controls and their effect. A good example of this is the
way signallers’ NX panels are laid out (for the most
part!) to mimic the track and signal layout
Page 31
Part 2: Guidance
Design
• Affordances – using the shape and other characteristics
of an object to suggest how it can be used. For
example, buttons ‘afford’ pushing and pulling, knobs and
switches ‘afford’ turning, ‘slots’ afford the insertion of
suitable objects
• Constraints – limiting the way an object can be used
and so reducing the possibilities for making errors. For
example, the design of floppy disk drives only allows
the disks to be inserted in the correct way
• Conventions – employing design practices that have
become conventions because they work well, as in the
case of handles that are pulled to open objects. When
conventions are used, people can learn quickly how to
use the system. When conventions are broken, they
can become confused and frustrated.
• Environment – considering the environment in which
an item of equipment is to be used. This means, for
example, putting switches in positions where it won’t
be easy to activate them by mistake. It also means
protecting and supporting the human under severe
environmental demands (eg temperature, noise,
vibration) – but this can be over-done! (see Panel,
Learning to ignore alarms!).
Learning to ignore alarms!
AWS alarm activation occurs frequently when drivers
travel over magnets for signals in the opposite direction
on the Maidstone Relief Line. This is because the alarm
is not suppressed on the bidirectional line. It is possible
that drivers who keep being subjected to false AWS
activation will become desensitised to the sound
and they may learn to ignore the AWS alarm at this
location.
Page 32
• Workflow – considering the tasks that the users will
undertake with the equipment. Effective analysis here
will mean that the displays and controls necessary
for more frequent and more critical tasks are readily
available. For example, an important status display may
be shown by default, and emergency shut down will be
directly to hand.
• Workload – considering the capabilities and limitations
of the users, and designing the interface so that they
are never overloaded – either mentally or physically.
For example, excess mental workload can occur
with multiple alarms, or if an interface requires users’
attention to be in several places at once – especially
at critical times (see Panel, Too much to do!). Similarly,
excess physical workload will occur if equipment is
designed without proper regard to the way muscles
and joints work. Designs that require heavy tools to be
pushed against a surface, or held in a precise position
or tight grip, will quickly cause fatigue.
What makes controls easy to operate?
Several factors affect how easy – or difficult – a control is
to operate. Sanders & McCormick (1993) include these:
• How easy it is to identify the control. Shape, texture,
size, colour, location, activation method and labels all
help with distinguishing between different control
types. Identical labels should be placed above both the
control and its corresponding display.
• Direction of control movement and its relationship with
its corresponding display1. When a control is moved or
1. Expectations about the direction of movement of controls is not the
same in all cultures. For example ‘on’ in the US is reflected by a switch up
position – the reverse of the UK.
Too much to do!
Class 175 Drivers have reported that they are required
to answer alarms, communicate with passengers via
intercoms and operate doors, as well as concentrate
on driving and keeping to the line speed. On occasions,
Drivers have been distracted by the acknowledge
(ACK) alarm and, while trying to ignore a multitude of
‘active’ TMS (Train Management System) alarms, have
sometimes found it difficult to differentiate the in-cab
radio telephone. When TMS was first introduced, it
was not unknown for the ‘toilet full’ alarm to sound,
thereby providing an unnecessary driver distraction at
potentially safety-critical moments.
turned to the right, it should mean ‘more’ or ‘on’ and
its display pointer must move right over a round or
horizontal display.
• Control/response (C/R) ratio – the relationship between
the movement of a control device by the operator
and the movement of the system in response, often
fed back to the operator via a system display. Low
C/R ratio control devices are ‘sensitive’ in nature, in
that a very small movement of the control results in
a marked change (as in the case of helicopter pilot
controls). This can lead to operators ‘overshooting’
the precise location required. Conversely, high C/R
ratio devices are ‘insensitive’, requiring larger operator
movements.
• Resistance in the control – elastic, spring-loaded or in
some other form. This provides most of the feedback
for users of such electro-mechanical controls.
Understanding Human Factors/June 08
Part 2: Guidance
Design
• Deadspace – the amount of control movement around
the null, home or neutral position that does not make
the controlled system device move. In some control
devices, significant amounts of deadspace may affect
performance adversely. Deadspace is less important
with less sensitive C/R relationships.
• Location and arrangement of components. As far as
possible, displays should be placed close to the
controls that affect them. Components should be
grouped by function and those that are important
and/or frequently used should be in a prominent
location. Components should be positioned to reflect
commonly used sequences, and sequences should be
arranged logically eg from left to right.
Summarised from Sanders & McCormick (1993)
Some of these recommendations may seem obvious, but
you need to remember that long or monotonous work
can cause boredom and fatigue which, in turn, lead to
reduced alertness, fatigue and errors. It’s important to
have a logical layout of controls and displays that assumes
people will fall back on long-established habits.
‘Style guides are an
important resource for
interface designers. Does
your organisation have one?’
What makes interfaces easy to use?
It is often assumed that new technologies create an
operating environment to which people will adapt easily.
But this is not always the case, and designers of systems
involving human-computer interaction (HCI) have to
put a lot of work into getting the interface between
users and computers right. In working on this interface,
designers need to draw on several disciplines, particularly
psychology. Studies by psychologists of sensation,
perception, attention, memory and decision-making
provide valuable guidelines on:
• screen layout
• information grouping
• use of colour and highlighting in interface design
• use of animation and shading
• menu length
• depth and breadth trade-offs in menu design
• alarms design and layout2.
An important resource that many interface designers
rely on is a style guide that is used across the whole
organisation or industry. This ensures that all userequipment interfaces are internally consistent and
promote a consistent look and feel. It can also ensure
that design guidelines are developed for specific classes of
equipment where components may be relatively unique.
2. RSSB are developing definitive best practice on alarms arrangements for
drivers’ cabs – see RSSB website.
Understanding Human Factors/June 08
What should a style guide include?
An interface style guide should cover all components
of the user-equipment interface. Information about the
organisation and the target audience – such as language
preferences and colour conventions – should also be
taken into account. Remember both stakeholders and
equipment designers will need to approve it.
The guide should consist of design principles stated as
rules (eg colour codes for alarms or equipment status)
and examples of user-equipment design components
(display layout, window design, button shape and
appearance, and so on.). It should be used to check and
enforce compliance during the design process.
You need to structure human factors style guides in
accordance with existing standards and current style
guides approved elsewhere in the organisation (see Panel,
What should a style guide include?).
What’s the most important interface design issue?
The common theme of all interface guidance is helping
the user to cope with complexity. The diagram on page
34 emphasises what users need to accomplish through an
interface.
Essentially, users need to structure information at a high
enough level so that they can hold the situation in their
heads and reason about it. They can then decide what
they want to achieve and plan how they will do it. Finally,
they must use the interface to carry out their plans.
It follows that a successful interface is one that first allows
users to maintain an accurate moment-to-moment
mental picture of the situation (usually called ‘situation
Page 33
Part 2: Guidance
Design
Interfaces must help users filter complexity
awareness’). It then needs to support the actions the
user must take to exert control over the situation. This
has powerful design implications for the way information
is structured and controls/displays are grouped. It also
highlights how important it is for designers to employ a
user-centred design approach (page 25).
What is situation awareness?
Put simply, situation awareness means knowing what
is going on around you – and what is likely to go on
in the future. For example, a signaller needs to know
about current train positions and routes, and they
need to predict future states so as to avoid possible
conflicts. In operational terms, situation awareness means
continuously extracting environmental information,
integrating this information with previous knowledge
to form a coherent mental picture, and then using this
picture to direct further perception and to anticipate
future events.
Most signallers – and other people too, like air traffic
controllers – have experienced the feeling of suddenly
losing ‘the picture’ when their workload (page 125)
has become too much. An important way in which
this picture is used is to create expectations for future
events. This allows operators to develop plans and
manage the complexity of situations. It follows that when
these expectations are wrong – due to an inaccurate
or failing picture – there may be severe safety-critical
consequences. The design of interfaces with appropriate
displays, annotations and feedback can
do much to allow operators to develop ‘Situation
and maintain an accurate mental picture.
The two tables about situation awareness on page
35 give good clues to interface designers on how
their products can support this vital aspect of user
performance.
How do you measure situation awareness?
There are several measures of situation awareness, of
which SAGAT (Situation Awareness Global Assessment
Technique) is the most widely known and certainly the
best publicised. SAGAT involves a user and a simulation
of the task. Jeannot explains
that the simulation is frozen at
awareness is
randomly selected times and
what you need to know
the user is asked about their
Jeannot (2003) says that three different
perception of the situation at
so that you are never
levels of information processing are
that instant. SAGAT queries are
surprised.’
involved in good situation awareness:
on specific data or data criteria
Air Traffic Control Instructor
corresponding to the three
1 Perception. This is the first
levels of situation awareness
fundamental step in situation awareness and involves
described earlier. The screen and all information sources
perceiving and attending to important cues in the
are blanked/hidden. Computerised versions of SAGAT
environment.
exist, but paper versions are probably more easy to use
(and modify). This technique could be used as part of
2 Comprehension. This step goes beyond mere
user training and appraisal programmes.
perception and involves integrating different pieces
of data and information and deciding on their
What challenge does situation awareness present?
operational relevance.
Interface designers need to work out how to present
3 Projection. This step involves being able to anticipate
relevant information so that it’s easier for users to
future events and their implications based on
integrate information from different sources (eg
comprehension of the environment. Projection allows
timetables, weekly notices, NX panel information) and
for timely decision making. It is what gives experts
using different channels (eg Concentrator, Panel, VDUs,
apparently lots of time to carry out their tasks when
talking to colleagues). Users need to understand the
compared to novices, who are much more ‘trapped’ in situation – not only what has been happening, but also
their immediate situation.
what this situation means for the immediate future.
Source: Jeannot et al (2003), reproduced with permission
Page 34
Understanding Human Factors/June 08
Part 2: Guidance
Design
Indications of good and bad situation awareness
Indications of good situation awareness
Indications of impaired situation awareness
•Anticipating events
•Increase in delay in communication response times
•Being able to predict the next task demand requiring attention
•Inconsistency in communications with colleagues
•Managing resources (technical system, team demands, communications
etc)
•Sudden and unexpected variation in workload
•Managing time
•Need to check the same information several times
•Confusion
•Feeling of being in control
•Taking the right decision at the best moment, eg managing traffic in a safe
and expeditious way
•Detecting mismatches and uncommon events
Source: Jeannot et al (2003), reproduced with permission
Factors affecting situation awareness, and strategies people use to recover it
Exclusively focusing on historical events without this
context (as TRUST does) simply transforms highly skilled
performance into a thankless task.
A particular design challenge is to support the user’s
need to project into the future, while preventing them
from acting on false expectations. As the discussion on
investigator biases revealed (Why do accidents happen?
page 18), people have a strong tendency to form
hypotheses and then try to confirm them. Interface
designs need to help users maintain an accurate
understanding of the big picture as much as they need to
help them operate according to what it tells them. There
are currently no rules on how to do this, apart from the
guidance provided by function allocation (page 37).
Factors leading to the loss of situation awareness
Strategies used to recover situation awareness
•Time pressure
•Increase delays between telephone calls and answering
•Focusing on non-pertinent, or less pertinent information
•Checking consistency between panel displays and timetable updates
•Focusing on a subset of relevant information, but missing the evolution of
other information
•Forcing oneself to speak slowly and precisely; return to strict phraseology
•Request help to decrease load
What are the design principles for effective interfaces?
•Becoming reactive rather than proactive
•Scrutinise all movements information more closely
•Reduced room to manoeuvre
•Force oneself not to spend too much time on a single problem
•Increased occurrence of non-safe situations
•Reorganise task strategy, eg change method of prioritisation
•Noise/distraction from other people
•Prioritise work and ‘forget’ less important tasks
•Mental and/or physical fatigue
•Always prioritise new information as it arrives
An interface design must connect the user’s purposes,
needs, capabilities and limitations to the task’s demands
for the key principles for interface designers (see Panel
on page 36, Ten principles for good interface design).
Remember – many design problems arise from poor
communication between the design team and the target
users, so make sure you use a user-centred design process
from the outset (page 25).
•Volume of train movements; sudden variation in traffic load
•Number of phone calls
•Lack of timetable/head code information at the right time
•Lack of adequate feedback
•Too much happening and having to process too much information
Wilson and Rajan (see Further information) give a very
comprehensive checklist of the interface design factors
that need to be taken into account.
•Traffic building up
•Unusual or unexpected events, eg unexpected communications with
colleagues, suddenly degraded working
Expert users deal with the present by understanding how
current events might develop in the future – and taking
specific action now to simplify things later. It is worth
noting in passing that the main reason that the TRUST
system is unpopular with many signallers is because it is
Understanding Human Factors/June 08
only concerned with historical events. Unlike signallers,
it is blind to all the much more undesirable events that
the signallers manage to avoid through effective action.
The results that skilled operators achieve can only be
fully appreciated in terms of what could have been.
Page 35
Part 2: Guidance
Design
Ten principles for good interface design
1 Ensure consistency and adhere to agreed standards
and conventions
2 Structure tasks and information intuitively and reflect
the real world experience of the job holder
3 Support user control (make sure the user feels in
charge)
4 Provide user help
RSSB is developing detailed human factors guidance on
alarm design and layout. In addition, HSE has published
fact sheet guidance on alarm handling – see Further
information. Further work by RSSB has resulted in a
CCTV toolkit which aims to ensure that the system
design can meet a variety of industry needs and support
operator capabilities.
Tools and techniques
5 Reduce short-term memory load (eg support
recognition as opposed to recall)
• Design Scenario Analysis
6 Provide visibility of system status
• Heuristic Analysis
7 Provide informative feedback for all user actions
8 Provide simple error handling, prevention, recognition
and recovery – allow easy reversal of actions (undo)
9 Design dialogues to yield closure (ie to ensure a task is
completed)
10 Provide short-cuts for frequent/expert users
Specific equipment interface guidance for the railways
There is a Railway Group Standard that sets out the
requirements for external visibility from inside driving
cabs for control facilities and other interior arrangements.
This Standard is designed to ensure a working
environment in which drivers of traction and rolling
stock vehicles and on-track machines can carry out their
operational duties safely and effectively.
In 2004, RSSB published human factors guidance relating
to GSM-R cab design. This provides a set of human
factors principles for the human-machine interface within
the cab. It also recommends an analysis technique based
on software manikins to assess alternative cab fitment
locations against human factors principles.
Page 36
• Layout Analysis
• Link Analysis
• Keystroke Level Model
See Usability testing on page 28 for recommended
evaluation techniques.
Further information on equipment/interface design
1 Bailey R.W. (1982) Human Performance Engineering:
A Guide for System Designers, Prentice-Hall, NJ.
© European Organisation for the Safety of Air
Navigation EUROCONTROL
5 Kroemer K.H.E. & Grandjean E. (1997) Fitting the Task
to the Human (5th Ed), Taylor & Francis
6 Pheasant S. (2001) Bodyspace. Anthropometry,
Ergonomics and the Design of Work. 2nd Edition,
Taylor & Francis
7 Railway Group Standard GM/RT2161 (1995)
Requirements for Driving Cabs of Railway Vehicles,
Aug 1995
8 Salvendy G. (1997) Ed. Handbook of Human Factors
and Ergonomics 2nd Ed. Wiley-Interscience Publishers
9 Sanders M.S. & McCormick E.J. (1993) Human factors
in engineering and design (7th edn), McGraw Hill, NY
10 Wilson J. & Rajan J. (1995) Human-machine interfaces
for systems control. In Evaluation of Human Work: A
Practical Ergonomics Methodology, Eds J.R. Wilson and
E.N. Corlett, Taylor & Francis
11 Woodson W.E. Tillman B. & Tillman P. (1992) Human
Factors Design Handbook. 2nd Edition, McGraw-Hill
2 Boff K.R. & Lincoln J.E. (1988) Engineering Data
Compendium: Human Perception and Performance.
John Wiley
3 HSE (2000) Better Alarm Handling. Fact sheet
www.hse.gov.uk/pubns/chis6.pdf (as of May 2008)
4 Jeannot E. Kelly C. & Thompson D. (2003) The
development of Situation Awareness measures in
ATM systems. EATMP report. HRS/HSP-005-REP-01.
Understanding Human Factors/June 08
Part 2: Guidance
Design
Function allocation
What is function allocation?
• between people and machines – where newer
dynamic approaches are taking over from the more
traditional fixed approaches used in the earlier days of
computers and before.
When designing equipment for use by human operators,
it’s often necessary to decide what parts of the system (ie
the equipment or the operator)
How is function allocation
‘The basic problem is that for
are going to do what. This is
done?
a system to be operated safely,
known as function allocation.
There are three main strategies
Function allocation may be fixed,
an intelligent human operator is
for fixed function allocation
where each system component
necessary. However, the human
between people and machines:
remains dedicated to the role it
operator
is
an
intrinsically
unsafe
is given. Alternatively, allocation
• the capability strategy
may be dynamic, meaning that
component of the humanit is designed to be done during
• the automation strategy
machine system.’ Maarten Boasson,
‘live’ operational time. Quite often,
Professor of Computer Science
• the economic strategy.
system and job designers are able
to arrange for a mix of fixed and
The capability strategy
dynamic functional allocation approaches.
This involves assigning each function to the most
capable agent (either human or machine). This is the
In broad terms, function allocation can take place:
traditional approach (see Panel, Fitt’s list) and relies on
• between people – where it is called ‘teamwork’ (page
what we know of the differences between human and
103) and is usually a mix of fixed (role-based) and
machine capabilities. For example, humans are better
dynamic function allocation
able to sense unusual and unexpected events, generalise
from observations, develop entirely new solutions
• between machines – where it may be fixed, with each
and detect stimuli when there is noise or clutter. By
machine programmed to fulfil its own – and only
contrast, machines are better at monitoring for preits own – function; or else
specified events, storing and
dynamic, with a redundant
‘The more advanced a control retrieving information, exerting
system component
system is, so the more crucial considerable physical force and
taking over from another
performing repetitive activities
component that has
may
be
the
contribution
of
the
accurately, rapidly and reliably.
overloaded or failed
human operator.’
Lisanne Bainbridge, Professor of Psychology
Understanding Human Factors/June 08
One problem with this strategy
is that machines and software
Fitt’s List – a traditional approach
In 1951, Paul Fitt concluded that humans appear to
surpass machines with respect to:
• Ability to detect small amounts of visual or acoustic
energy
• Ability to perceive patterns of light or sound
• Ability to improvise and use flexible procedures
• Ability to store very large amounts of information
for long periods and to recall relevant facts at the
appropriate time
• Ability to reason inductively and exercise judgement
And in 1951, machines appeared to surpass humans
with respect to the following:
• Ability to respond quickly to control signals, and to
apply great force smoothly and precisely
• Ability to perform repetitive, routine tasks
• Ability to store information briefly and then to erase
it completely
• Ability to reason deductively, including computational
ability
• Ability to handle complex operations, ie to do many
different things at once
Summarised from Fitt et al (1951)
are evolving all the time and what is true at the design
stage may not be true when the system is fielded, let
alone later in its life. A second problem is that functional
allocation is rarely left entirely to the designer. In practice,
Page 37
Part 2: Guidance
Design
design decision making may be severely constrained by
earlier management decisions (eg based on policy or
union agreements) and system considerations (eg the
need to fit with legacy systems – ie older systems still in
use). This first strategy has been used a great deal, but is
of limited value and is no longer recommended.
Ironies of automation
1 Designers may assume that humans should not be
included in systems because they are unreliable and
inefficient. But many operating problems come from
designer errors.
The automation strategy
2 Automated systems are implemented because
they can perform better than the operator, yet the
operator is expected to monitor their progress.
This involves allocating to the machine every function
that can be automated. It is favoured by many engineers
as a way of reducing the risk of error, but it is dangerous
in practice. First, not everything can be automated, and
leaving unrelated bits and pieces of functions to people
leads to boredom, inattention, de-motivation and overall
poor performance. Second, machines are not perfectly
reliable, and the more complex they are, the more they
fail. This usually means employing higher skilled, better
trained and more expensive operators, who – with
not enough to do – suffer from even higher levels of
boredom etc. This strategy has been used often, but is
also of limited value and is not recommended (see Panel,
Ironies of automation).
The economic strategy
This involves finding an allocation scheme that ensures
economical efficiency. Here, designers start by allocating
functions that clearly must be done by humans or
machines – for mandatory organisational, political
or safety reasons. They then distribute the remaining
(probably most) functions between humans and machines
in a number of alternative configurations. The actual
configuration finally chosen needs to be an economic
trade-off between the following factors:
3 The operator is reduced mostly to monitoring,
leading to fatigue and ineffectiveness.
4 In the long term, automation reduces physical and
mental skills of workers, yet these skills are still
required when automation fails. In fact, the skills may
be in more demand than usual when automation
fails, because there is likely to be something wrong,
requiring, takeover, diagnosis and recovery.
De-skilling also affects workers’ attitudes and health
adversely.
Adapted from Bainbridge (1987), reproduced with permission
• developmental and operational costs
• maintainability over the planned system lifetime
• portability to different departments or organisations
• producibility (ie how easily they can be manufactured)
• safety
• staffing and training costs
• ability to meet defined user requirements
• ability to satisfy human needs.
Page 38
This third, economic, strategy for functional allocation is
the one recommended in this Guide, as part of a usercentred design process (page 25).
What are the rules for fixed function allocation?
The golden rule to remember is that machines must be
made to support human purposes, not the other way
round. There are three other key rules.
1 Allocate all clear-cut functions – where there are
mandatory reasons for allocating a function, or
portions of it – to either humans or machines.
2 Locate all remaining functions in a matrix that sets
human performance against machine performance
(both axes should use a simple scale of Unsatisfactory
to Excellent) (see Panel on page 39, What are our
basic mental capacities and limitations?).
If a function is in the unacceptable region of the
matrix (it cannot be satisfactorily performed by either
humans or machines), then the function must be redefined or the system requirements must be modified.
If the function is shown to be unacceptable either for
humans or machines then it should be treated as a
clear-cut function and assigned accordingly.
If a function is in the region that is better performed
by humans, then it can be tentatively allocated to
the human alternative. Correspondingly, if it is better
performed by machines, then it should be allocated
to the machine alternative. However, these allocations
may be changed by the third key rule (below).
Understanding Human Factors/June 08
Part 2: Guidance
Design
3 Refine the preferred allocations according to:
• Utility. A function may be allocated to humans
simply because their presence is required, or there
is a compelling reason why they should perform
the work eg it requires judgement
• Health and safety. A function may be allocated
to humans or machines to reduce the risk to
health and safety. The designer can prepare a
list of potential injuries (cuts, bruises, fractures,
amputations, burns, internal ruptures, eye
penetration, asphyxiation, etc) and hazardous risks
to health eg damage to sight, hearing, exposure to
noxious substances etc). This can then be used as a
checklist for human-machine allocation decisions.
• Cost. Consider the relative cost of human and
machine performance and allocate on the basis
of least cost. You will need to consider all the
aspects of cost that are relevant to you. These are
likely to include developmental and operational
costs; maintainability costs over the planned
system lifetime; costs of portability to different
departments or organisations; cost of manufacture;
and associated staffing and training costs.
• Emotional needs. A function may need to be
allocated to humans because they need to know
that their work is recognised for its value, to feel
personally secure, or to feel that they are in control.
• Information needs. A function may need to be
allocated to humans because they need information
in order to be ready for actions and decisions that
may be required.
Understanding Human Factors/June 08
What are our basic mental capacities and limitations?
Attention
•Attention on a task can only be sustained for a fairly short period of
time, depending on the complexity of the task. The usual figure cited
is around 20 minutes, after which fatigue sets in and errors are more
likely to occur.
•People can only pay attention to a small number of tasks at once.
For example, if a signaller is focused on handling a particular train,
then it is likely that they will be less attentive to other aspects of
safety, or other alarms.
procedures within a new task. However, trained individuals are able
to retain larger chunks of information in memory. For example, chess
grandmasters can remember the location of more pieces on a chessboard because they see the pieces not as single units, but as parts of
larger conceptual units
•Accessibility – even when items are stored in memory, it is sometimes
difficult to access them. People are much more likely to remember
information if they are in similar conditions to when they encoded
the information. For example, signalling staff trained in a classroom
may not be able to remember relevant details in a signal box.
•If a task is repeated often enough, we become able to do it without
•Levels of processing – another way in which information can be more
conscious supervision, although this ‘automation’ of repetitive
reliably remembered is to learn it at greater depth. For instance, if it
behaviour can force us into mistakes. In 1979, an operator at
is necessary to remember items from the Rule Book, then it helps to
Oyster Creek Nuclear Power Plant intended to close off two pump
understand more about the conceptual framework behind the rules.
discharge valves. Through an attentional slip, he accidentally closed off
If only the words that express the rule – rather than its meaning
two other valves as well, and in doing so, closed off all cooling to the
and purpose – are remembered, then there is a higher chance of
reactor core.
important information being forgotten.
Perception
•We do not have direct access to the world around us. Instead, we
must interpret information we sense. The more visual information
available to the perceiver, the less likely it is that errors will be made.
Bearing this in mind, systems that include redundant information in
their design may cause fewer accidents. An example of this was the
change in electrical earth wire colour coding in the 1970s to include
not only colour, but also a striped pattern. At the same time the live
wire was changed from red to brown to overcome problems caused
by red-green colour blindness, a condition which affects 1 in 10 of
the male population.
•The more intense a stimulus (such as a light or a noise), the more
powerful the response elicited (such as brain activity or a physical
movement). This has implications for the way danger signals and
alarms should be presented at work.
Memory
•Capacity – short-term memory has an extremely limited capacity. In
general, people can remember no more than around seven recently
presented individual items at a time. This has safety implications in
areas such as giving new workers a set of instructions to follow from
memory or attempting to remember the correct sequence of
Logical reasoning
•Humans are not very good at thinking logically, but in technological
situations, logical procedures are often necessary (for example,
troubleshooting an NX panel which has developed a fault).
•A common source of error in industry is the behaviour which arises
from design which may be logical in a formal sense, but which is
inappropriate to the way people think about their jobs - especially
when they are under stress.. During the Three Mile Island incident
in 1979, two valves which should have been open were blocked
shut. The operators incorrectly deduced that they were in fact open,
by making an assumption about the instrument display panel. The
display for the valves in question merely showed that they had been
instructed to be opened, whereas the operators took this feedback
as an indication that they were actually open. Following this, all other
signs of impending disaster were misinterpreted with reference to
the incorrect assumption, and many of the attempts to reduce the
danger were counterproductive, resulting in further core damage.
Adapted from Managing Human Error, POSTNOTE, June 2001, Parliamentary
Office of Science and Technology, reproduced with permission
Page 39
Part 2: Guidance
Design
Note
Where adequate information for allocation is not available,
human factors judgements by human factors experts based
on partial information will result in better design decisions.
When is function allocation done?
Function allocation is one of the first sorts of human
factors issues you need to consider when designing a
system. Once you have made the necessary decisions
about function allocation, you are ready to consider the
human factors topics of task analysis (page 47) and job
design (page 50).
Fixed function allocation
For fixed function allocation, all of the decisions are
necessarily taken fairly early in the design process.
However, while the decision to use dynamic function
allocation needs to happen early, the actual allocations
take place during operational activity.
Dynamic function allocation
Dynamic function allocation takes place all the time within
human teams. Here, people are able to understand what
they are trying to achieve as a team, what each other’s
responsibilities are and when to step in to help out other
team members. (See the section on teamworking, page
103). In more recent years, machines and computer
software have become sophisticated enough for
something along these lines to happen between humans
and their equipment. Known as adaptive automation, it
involves the control of functions shifting between people
and machines dynamically in response to environmental
factors, operator workload and performance.
Page 40
Human-triggered dynamic function allocation
Machine-triggered dynamic function allocation
At present, dynamic allocation is most commonly
human-triggered and involves the human operator
either engaging or disengaging some kind of automation.
A classic example is when an airline pilot engages/
disengages autopilot. An obvious rail example is when
an IECC signaller uses Automated Route Setting (ARS)
software.
Increasingly, dynamic allocation is being arranged via
machine or software triggers. One example is the Airbus’s
ability to prevent the human crew from putting the
aeroplane into a stall.
Factors that influence when the operator will engage
automation include:
• their own current state – the extent to which their
workload (page 125) and ability allows them to
perform the function that could be allocated to the
machine
• the cost of allocation – the adjustments they will have
to make (eg extra communications) in order to
allocate the function to the machine
• the machine’s ability – whether the machine can carry
out the function, given the current situation and
problem conditions (eg whether the machine has
enough processing capability)
• confidence in the machine – whether the machine will
perform the function well enough.
‘Dynamic function allocation
takes place all the time
between humans – where it’s
called teamwork.’
Railway examples (in increasing order of sophistication)
include the Automatic Warning System (AWS), the Train
Protection and Warning System (TPWS), Automatic
Train Protection (ATP) and the European Rail Traffic
Management System (ERTMS).
All of these machine-triggered allocation systems use
measure-based triggers, ie machines take over functions
on the basis of an automated detection or assessment of
procedure violations or human workload. An alternative
approach, which is presently only at the research stage,
is the model-based trigger. The idea here is that an
automated system is able to recognise or predict human
states that indicate the need for function re-allocation.
The system then shifts tasks away from the affected
operator, either to another human or to the system.
Highly capable automated systems can make important
contributions to system safety, precision and efficiency.
But they also impose system costs in the form of new
opportunities for error, which occur when humanmachine coordination breaks down. The human operator
is increasingly responsible for ensuring cooperation and
resolving conflict between human and machine intentions
and actions. However, if you dig a little deeper, you find
that many of these joint systems only perform adequately
because the human agents are resourceful and adaptable
in the face of uncommunicative and uncooperative
Understanding Human Factors/June 08
Part 2: Guidance
Design
machines. Time pressure, workload, and problems with
situation awareness can reduce the operator’s ability to
coordinate properly with others.
Users need to be able to see what automated agents
are doing and what they will do next. Users also need
to be able to re-direct machine activities fluently when
they recognise a need to intervene. Under manual
control, human operators often obtain enough feedback
about the results of their actions within a few seconds to
correct their own errors. But there are many examples
of human operators making the same types of error in
setting up and monitoring automatic equipment that does
not give adequate feedback. The design needs to take this
into account and, for example, provide displays to help
operators who have been interrupted in mid-sequence.
These problems must be addressed with an appropriate
mix of equipment design (page 31), job design (page 50)
and training (page 55).
Workplace design
How important is it?
Designing the work environment or workplace to
meet users’ needs is as important as the appropriate
design of equipment and human-machine interfaces.
Poorly designed work environments can have serious
consequences, such as schedule delays, recurring
discomfort, a decrease in performance or output, and
possibly permanent injury or death. Human beings are
remarkably adaptable and we can adjust to whatever task,
furniture or equipment we are given – but we sometimes
do so at great cost. Problems such as Repetitive Strain
Injury (RSI), for example, are often a specific result of the
body adapting to a repetitive task or an inappropriate
posture.
What are the principles of workplace design?
Further information on function allocation
The overall aim of workplace design is to create a
comfortable and stress- and hazard-free environment,
which is compatible with performing tasks over time.
More specifically, it aims to:
1 Bailey R.W. (1982) Human Performance Engineering:
A Guide for System Designers, Prentice-Hall, NJ
• decrease the number and cost of accidents, injuries
and disabilities
2 BS EN 614-2:2000. Safety of machinery. Ergonomic
design principles. Interaction between the design of
machinery and work tasks
• improve the responsiveness of the organisation as a
whole to its customers
3 Bainbridge L. (1987) New Technology and Human
Error, chapter 24, 271-283. John Wiley
4 ISO 10075-1:2000. Ergonomic principles related to
mental workload: General terms and definitions
Understanding Human Factors/June 08
• ensure that organisational systems work as well as
possible
• decrease physical and mental stress on personnel
• increase job satisfaction and productivity.
Four principles of workplace design
Four principles govern the broad arrangements of
workplace design. They may seem like common sense,
but are often overlooked.
1 Importance. Components that are essential to
safe and efficient operation should be in the most
accessible positions. ‘Accessible’ refers not just to
ease of reach, but also to visibility, audibility etc.
2 Frequency of use. Components that are used
frequently should be the most accessible.
3 Function. Components with closely related functions
should be located close to each other.
4 Sequence of use. Components that are often used in
sequence should be located close to each other and
their layout should be consistent with the sequence
of operation.
Source: Pheasant, in Wilson & Corlett (1995) (following original work by
Sanders & McCormick,1993), reproduced with permission
There is extensive information about our physical
capabilities and limitations; about typical (and not so
typical!) body measurements; and about likely user
perceptions and behaviour (see Further information at
the end of this section). Workplace designers can draw
on this information in focusing on comfort, performance
and health & safety. These criteria are related: case
studies show that increased comfort and well being
will improve productivity (and vice versa). The amount
of improvement is typically in the order of 25% (not
counting the longer term value of reduced sickness due
to workplace conditions such as back and neck pain,
and RSI). Many aspects of the three criteria of comfort,
performance and health & safety should be served well if
Page 41
Part 2: Guidance
Design
you pay attention to four key design principles (see Panel
on page 41, Four principles of workplace design).
How is workplace design best approached?
Once you have analysed the relevant user-centred design
data (page 25) and developed the initial concept, it
is good practice to create a succession of mockedup workplace prototypes. You then need to evaluate
these under representative conditions using typical user
subjects. User trials should be designed to investigate the
spatial arrangements for the avoidance of discomfort,
inadvertent hazard potential etc. Observations should
be recorded and appropriate design modifications made
where necessary.
You can create preliminary, reduced-scale mock-ups
with foam or toy bricks. At later stages, it’s better
to use full-scale wooden models. Workspace and
clearance estimates should be assessed using a range of
representatives of the user population wearing their likely
clothing. More conveniently, precision graphical CAD
environments using manikins, eg SAMMIECAD, can be
used.
It’s important to note that documentation of human
factors during the design process is becoming more
urgent as legal aspects of design-induced injuries place
the burden of safety on the designer as well as the
manufacturer.
What should be taken into account?
In designing a workplace that meets human factor
requirements, designers need to take the following factors
into account.
Page 42
Analysing body size and postural requirements
Traditionally, body size and postural considerations have been investigated via a fitting trial in which an experiment is
conducted with a range of possible users and an adjustable prototype. The aim is to investigate where the ‘just right’ point
is for the majority of users. An alternative approach is to use the method of limits technique. This is a pencil and paper
technique that uses body size data to predict what the results of a fitting trial would have been had it been performed.
A third technique involves the use of simulation and modelling. Here, body size data drives software models which
are allowed to interact to produce overall task timings or error rates (eg as in the Integrated Performance Modelling
Environment - IPME) or else as three-dimensional manikins that can move around within an accurate simulation of
the prototyped workplace (eg SAMMIECAD). Whichever technique is used, body size and postural considerations are
generally addressed with three different sorts of measurement:
•clearance (eg head room, knee room, elbow room etc). Here the limiting user will (usually) fall into the largest 5% of the
population (referred to as the 95th percentile)
•reach (eg location of controls with respect to the seat). Here the limiting user will (usually) be someone who falls into
the smallest 5% of the population (referred to as the 5th percentile)
•posture (eg height of working surface which supports a particular task such as manual assembly or computer work). Here
the limiting user will be the average user ( the 50th percentile).
For each category, user adjustability of workplace components (eg seats, desk height etc) should be built around the 5th
and 95th percentile points.
Based on work by Pheasant, in Wilson & Corlett (1995), reproduced with permission
Body size
Manufacturers of furniture and equipment sometimes
base their designs on the measurements of the ‘average’
user. But more often they design with all sizes in mind,
from the smallest to the largest. For example, the height
of storage space might be set so that 90% of a typical
office population can reach it, or a doorway might be
designed so that 99% of the entire population can pass
through it without stooping (see Panel, Analysing body size
and postural requirements).
Postural requirements
The structure and arrangement of furniture and
equipment should not only help people to maintain a
healthy posture, but also allow them to vary or alternate
pressure points and body positions at will. It’s a good idea
to provide spaces for people to walk short distances,
adjust their chairs or alternate between seated and
standing tasks.
Personal space
People need enough space to move about and perform
various tasks. They also need ‘personal space’ (into which
they allow only people with whom they want to have a
personal discussion), ‘social space’ (in which they expect
to make purely social, temporary contacts), and ‘public
space’ (in which they don’t expect to have direct contact
with others). They can become tense or anxious when
Understanding Human Factors/June 08
Part 2: Guidance
Design
Guidance on lighting
•Avoid lighting levels that are too low or too high for the
amount of detail or contrast of the task – they will cause the
operating mechanisms of the eye to operate at their limits,
causing visual fatigue.
•Prevent glare and surface reflections wherever possible.
Reflections can obscure parts of the task that are necessary
for effective performance, and can often lead to headaches
or postural discomfort. In interior environments, glare is most
often caused by sunlight or improperly diffused artificial lighting.
•Prevent flicker wherever possible. Flicker (for example, from
artificial lighting or reflected movement) causes discomfort, and
usually causes distraction from the task at hand.
•Avoid wide variations in lighting levels across the working area.
If two or more parts of the task have very different lighting
levels, the eyes are forced to adapt continually between the
different levels, causing visual fatigue and discomfort. Large
contrasts in the user’s environment have a similar effect (eg
a daylight window directly behind a display screen can cause
discomfort as the eyes adapt between the two. For interior
environments, lighting ratios no greater than 5:1 should be
used. In an office where, for example, the task illuminance is
500 lux, the minimum illuminance in adjacent areas should be
no less than 100 lux.
•Aim for lighting ratios for task-to-immediate surround of
3:1. Task performance is reduced and concentration is more
difficult to maintain if the task lighting is lower than the
immediate surround.
•To aid maintenance, lift-off back panels should be completely
removable to reveal light-coloured interiors designed to assist
the spread of light within.
•Ensure that emergency lighting is powered by an independent
source, will be immediately effective and will provide sufficient
light to enable people to do what is necessary to ensure their
health and safety.
Reproduced from DEFSTAN 00-25 Part 19, with permission
© Crown copyright material is reproduced with the permission of the
Controller of HMSO and the Queen’s Printer for Scotland
Understanding Human Factors/June 08
the boundaries between these
different types of space are
breached. These boundaries are
best confirmed in user trials.
Shape (generally proportion)
The psychological effects of colour in the workplace
Colour
Distance effect
Temperature effect
Mental effect
Red
Closer
Warm
Very stimulating, Danger! Stop!
Orange
Much closer
Very warm
Exciting, Attention! Look out!
Green
Further away
Cold to neutral
Very restful, Exit, OK
Blue
Further away
Cold
Restful, Information available
If a space is out of proportion
Yellow
Closer
Very warm
Exciting
(too narrow, wide, high, etc),
Brown
Much closer
Neutral
Restful
people will consider it distracting
Violet
Much closer
Cold
Aggressive, tiring
or oppressive. If the space
Developed from Kroemer & Grandjean (1997), reproduced with permission
contains distortions – such
as curved surfaces, acute wall
• general and specific lighting design criteria for a range
junctures, and too many projections or surface changes
of interior and exterior applications
– people will consider it confusing and difficult to
• the cost-effective use of energy.
manoeuvre in.
Lighting
A space that is too dark tends to make people less active,
or feel anxious. A space that is too bright may make
them feel overly exposed, or they will complain of glare
or thermal discomfort. The only way of deciding how to
light a workplace is to test it with observers in tasks and
conditions equivalent to the operational environment.
The main British Code of Practice on lighting is published
by the Lighting Division of the CIBSE. The CIBSE Code for
Interior Lighting (1994) gives more detailed guidance (see
Further information) and should be consulted for further
information on:
• the effect of lighting conditions on the performance of
tasks in a wide range of interiors
• the appearance of an interior
Lighting design in both interior and exterior environments
should aid:
• task performance, by providing enough light to make
details of the task easy to see, so enabling high levels of
speed and accuracy
• safety, by allowing people enough light to see hazards
or potential hazards
• visual comfort, by providing the light needed for
comfortable and effective working (see Panel, Guidance
on lighting).
Colour
Strong colour can be a distraction, but can also have
attention-holding properties and help people locate
critical information. Generally, the tones chosen for a
workstation should be neither excessively bright nor
excessively dark. Different colours have particular
Page 43
Part 2: Guidance
Design
and then extends gradually to the
psychological effects. Kroemer &
‘The choice of colours,
lower frequencies. In addition to
Grandjean (1997) say that broadly
speaking, all dark colours tend to be tones and contrasts within physical damage, noise may result in
oppressive and tiring: they absorb
a workstation is of greater impaired alertness, disturbed sleep,
increased stress levels and irritability.
light and are difficult to keep clean.
operational importance
All light colours tend to be bright,
A special problem on the railways
friendly and cheerful: they scatter
than you might expect.’
is that the ambient noise in many
more light, and encourage greater
environments (eg track, driver’s cab)
cleanliness. Some colours transmit
strong safety messages (see Panel, The psychological effects and communication channels (eg radio, mobile phone)
can make it difficult to understand what people say (see
of colour in the workplace).
Why is communication so difficult?, page 106).
Noise
Kroemer & Grandjean (1997) say that noise is any
unwanted sound. Disturbing noise may arise from the job
in hand (eg track machinery, rolling stock, engine noise)
or from external sources (eg general office noise, general
depot noise). Some human activities are much more
noise-sensitive than others (eg concentrated mental work
or tasks where understanding of speech is important vs
hand signalling or some trackwork). Some people are
much more sensitive to noise than others. This sensitivity
does not just affect people’s comfort levels, but can also
make physiological damage more likely.
High noise levels experienced over a period of time can
lead to hearing impairment or loss. High frequencies are
more damaging than low frequencies. Intermittent sounds
like hammering are more damaging than continuous
noise. A single very loud noise, such as a shot or
explosion, can cause instant permanent damage. Usually
loss is only temporary at first, but the more often it is
repeated, the more permanent the damage is. Hearing
loss tends to occur in the mid-range frequencies first,
Page 44
ambient sound by generating exactly the opposite sound
at the same levels (so called ‘anti-noise’), while boosting
important sounds such as speech or alarms. Since 2003,
European Regulations have been in effect to control
exposure to noise at work. EC Directive 2003/10/EC
reduces limits on personal noise exposure to a limit value
of 85 dB(A) and an action value of 80 dB(A).
Vibration
Kroemer & Grandjean (1997) say that vibration is felt
as an imposition and a burden, ranging from minor,
to unbearable annoyance. The extent of the nuisance
depends on several factors, of which frequency is the
Designers can deal with noise by planning for no noise.
most important. In the short term, unacceptable levels
Methods include:
of vibration cause breathing problems, pains in the chest
and vital organs, backache, nausea and vomiting. In the
• using noiseless components and tools
longer term, vibration can cause degenerative changes
in the spine (from power tools), prostate and intestinal
• re-locating noise emitting equipment
ailments, arthritis, bone atrophy and ‘dead finger’. Spinal
• reducing the noise at source, eg through engine baffles damage tends to occur in people who suffer vertical
or using less noisy materials
vibration – usually via seated positions. Finger, hand and
arm problems tend to occur in people using power tools.
• reducing noise propagation, eg sound insulation or
The effects of vibration on human performance and
using buffer rooms between
comfort are well understood.
noisy and quiet places
‘A special problem on the
Designers can address vibration
• paying attention to personal
railways is making sure that issues by paying attention to
suspension systems and seating
noise protection, eg wearing ear
critical
communications
are
in vehicles, and by engineering
plugs or defenders.
damping solutions in power tools.
heard.’
EC Directive 2003/10/EC places
Unfortunately, many workers object
limits on worker’s exposure to whole-body and handto personal ear protection because it might mean that
transmitted vibration.
they miss some vital information from the environment
or they cannot hear other people speaking. One
possibility here is to use active defenders. These can block
Understanding Human Factors/June 08
Part 2: Guidance
Design
Indoor climate
As most people prefer to work How to maximise comfort for indoor workers
in natural daylight, it’s important Sedentary workers
Indoor climate refers to the temperature of the air and
Manual workers working in hot environments
to make full use of it wherever
surrounding surfaces, plus humidity, air movement and
•Ensure workers can acclimatise to heat in stages. New or returning
•Ensure the air temperature in
possible. At the same time,
air quality. The effects of all of these factors on human
workers should start by spending 50% of the working time in the heat
winter is between 20ºC and
and then increase by 10% per day.
21ºC and in summer between
you need to ensure that the
performance and comfort are well understood and
20ºC and 24ºC.
•Ensure that cooling periods increase with physical effort and/or heat
use of daylight does not cause
controllable for a wide range of activities, from sedentary
load.
•Ensure surface temperatures
discomfort through glare or
through to heavy indoor work (see Panel, How to
are within 3ºC of the air.
•Ensure that workers have the opportunity to drink little and often. A
heat gains for other users of an •Ensure the relative humidity
maximise comfort for indoor workers).
cupful every 10-15 minutes is recommended.
interior space. The Workplace
does not fall below 30% in
•Encourage the consumption of plain water with only occasional tea or
winter and does not move
coffee. Lukewarm drinks are absorbed faster than cold drinks.
The most common problems for people working indoors (Health, Safety and Welfare)
outside 40% - 60% in summer.
•
Encourage workers to avoid iced drinks, fruit juices and milk-based
Regulations 1992 state that
arise from the combination of physical work and air
•Ensure draughts between the
drinks since they put more stress on the digestive organs.
every workplace shall have
temperature. Kroemer & Grandjean (1997) say that
head and knees do not exceed
•
Ensure that drinking water is always available close to the workplace
suitable and sufficient lighting,
overheating leads to weariness and sleepiness, reduced
0.2 metres/second
and that workers can drink whenever needed.
and, so far as is reasonably
physical performance and increased liability to make
•Where radiant heat is excessive, eg from industrial heat sources,
practicable, lighting should
errors. Overcooling induces restlessness, which in turn
ensure the provision of eye protection, screens and protective clothing.
Source: Kroemer & Grandjean
be by natural light. In most
reduces alertness and concentration – particularly on
(1997), reproduced with
•Ensure everything is done to reduce the impact of heat on workers,
permission
eg through increased ventilation and/or dehumidification.
circumstances, suitable and
mental tasks. Sedentary workers are vulnerable to both
sufficient
lighting
for
both
overheating and overcooling. For obvious reasons, manual
directly observed. Generally, they like to sit where the
interior and exterior environments can be provided by a
workers tend to be more vulnerable to overheating.
entrance to the room is still within their line of sight and
combination of natural and artificial lighting.
will seek out a seat that is not next to an occupied one.
Windows
These sorts of issue have implications for the design of
Attractiveness
Generally, most people don’t like to live and work in a
open plan areas.
Attractive surroundings not only help people to feel
space that does not have windows. They seem to need
comfortable and in control, but also
some sort of contact with the
Circulation
‘The
most
common
improve their self-esteem. It has
outside world to feel safe. On the
In effective circulation spaces, people can move efficiently
even been found that a luxurious
other hand, too many, or oversized,
from place to place. The most efficient routes are in
physical problems for
setting has the psychological effect
windows can make people feel
a straight line and offer an unimpeded view of the
people working indoors
of making people speak more
vulnerable and anxious. Windows
destination. If a route is too complicated, people will react
quietly!
are also a key source of natural light,
to it with frustration and instinctively resist using it.
arise from the combination
which promotes visual comfort as
For information specific to the design of train driver cabs
of physical work and air
well as being useful for tasks where
Proximity to others
(see Panel, Cab design guidance for the railways).
good colour rendering is important.
People
enjoy
watching
other
temperature.’
people, but they do not like to be
in situations where they can be
Understanding Human Factors/June 08
Page 45
Part 2: Guidance
Design
Cab design guidance for the railways
The 1995 Railway Group Standard sets out the
requirements for external visibility from inside driving
cabs for control facilities and for other interior
arrangements (see Further information).
Tools and techniques
• Cab environment (heating, ventilation, air
conditioning, noise, toilet facility, and vibration)
SAMMIECAD is a computer-based human modelling
tool. As of May 2008, it is fully described at www.lboro.
ac.uk/departments/cd/docs_dandt/research/ergonomics/
sammie/home.htm The system allows designers to carry
out a 3D analysis of fit, reach, vision and posture. It runs
on the Windows NT/2000/XP platform and is particularly
appropriate for the design and layout of equipment and
furniture in public areas, offices and homes; cockpit, cabin
and interior evaluations for all types of vehicles; design
of control panels; field of view, reflection and mirror
evaluations; and safety and maintenance evaluations.
• Cab layout (general design, access, visibility and
seating)
We are grateful to Taylor & Francis for permission to reproduce portions of
Kroemer & Grandjean (1997) in the foregoing section.
In 1998, the US Department of Transportation
published comprehensive human factors guidelines for
locomotive cabs. The key topics covered include:
• Workstation design (controls, electromechanical
displays, auditory devices, general principles,
automation, electronic displays and computer input
devices).
Summarised from: Multer et al (1998)
In 2002, the International Union of Railways revised
detailed guidance on the ergonomic layout of train
drivers’ cabs. The topics covered include:
• Number and arrangement of seats for drivers and
other staff
• Dimensions and layout of the driver’s cab
• Visibility from the driver’s cab
• Driver’s desk and main operating equipment and
control systems.
Summarised from: International Union of Railways (2002)
Page 46
Further information on workplace design
1 Boff K.R. & Lincoln J.E. (1988) Engineering Data
Compendium: Human Perception and Performance,
John Wiley
2 BS 5940 Part 1 (1980) Design and dimensions of
office workstations, desks, tables and chairs
3 BS EN ISO 11064-1:2001 Erg. Design of Control
Centres. Principles for the Design of Control Centres
4 BS EN ISO 11064-2:2001 Erg. Design of Control
Centres. Principles for the Arrangement of Controls
5 BS EN ISO 11064-3:2001 Erg. Design of Control
Centres. Control Room Layout
6 BS EN 349:1993. Safety of machinery. Minimum gaps
to avoid crushing of parts of the human body
7 BS EN 547-1:1997. Safety of machinery. Human
body measurements. Principles for determining the
dimensions required for openings for whole body
access into machinery
8 BS EN 547-2:1997. Safety of machinery. Human
body measurements. Principles for determining the
dimensions required for access openings
9 BS EN 60598-1:2000. Luminaires. General
Requirements and Tests
10 BS 5266-1:1999. Emergency Lighting. Code of Practice
for the emergency lighting of premises other than
cinemas and other premises used for entertainment
11 CIBSE Code for Interior Lighting (1994) Chartered
Institution of Building Service Engineers
12 CIBSE Lighting Guide 3. (LG3) (1996) The visual
environment for display screen equipment use
13 EC Directive 2003/10/EC (2003) on the exposure of
workers to the risks arising from physical agents (noise
and vibration)
14 International Union of Railways (2002) Layout of
driver’s cabs in locomotives, railcars, multiple-unit
trains and driving trailers, UIC leaflet 651, 4th edition
15 Kroemer K.H.E. & Grandjean E. (1997) Fitting the
Task to the Human: A Textbook of Occupational
Ergonomics, Taylor & Francis
16 Leather P. Pyrgas M. Beale D. & Lawrence D. (1998)
Windows in the workplace: Sunlight, view and
occupational stress. Env. & Behavior, 30, 739-762
Understanding Human Factors/June 08
Part 2: Guidance
Design
17 Multer J. Rudich R. & Yearwood K. (1998) Human
Factors Guidelines for Locomotive Cabs, U.S. Dept of
Transportation Federal Railroad Administration, Office
of R&D, 400 7th Street, S.W. Washington, DC 20590
18 Osborne D.J. (1987) Ergonomics at work. (2nd Ed)
Wiley-Interscience
19 Pheasant S.T. (1991) Ergonomics, Work and Health,
Macmillan
20 Pheasant S.T. (2001) Bodyspace. Anthropometry,
Ergonomics and the Design of Work. 2nd Edition.
Taylor & Francis
21 PP 7317. Ergonomics Standards & Guidelines for
Designers, by Stephen Pheasant, BSI
Task analysis
What is task analysis?
Task analysis is not one specific technique, but rather a
methodology for collecting and recording information
about tasks in a systematic way. You can use it whenever
you need to understand exactly what a task involves
in terms of the knowledge, skills and behavioural
requirements of people. Task analysis is particularly
important in three main areas of organisational activity:
• Design – where it can help with the specification of
interfaces (page 33) and workplace arrangements (page
41), as well as function allocation (page 37) between
users and machines
22 Railway Group Standard GM/RT2161 (1995)
Requirements for Driving Cabs of Railway Vehicles
• Training – where it can help determine what needs
to be trained, as well as the performance standards
against which people must be assessed
23 Salvendy G. (1997) Ed. Handbook of Human Factors
and Ergonomics 2nd Ed. Wiley-Interscience
• Staffing – where it can help inform accurate recruiting
needs and selection criteria.
24 Sanders M.S. & McCormick E.J. (1993) Human factors
in engineering and design (7th edn), McGraw Hill, NY
There are several techniques for doing task analysis. The
most developed is Hierarchical Task Analysis (HTA), which
arose out of the need to understand the components
of complex, real world tasks in safety-critical industries
such as the chemical processing industry. In HTA, system
goals and sub-goals – and the activities needed to achieve
these goals – are described in more and more detail until
it’s possible to make design decisions at the lowest level.
HTA is a traditional form of task analysis which focuses on
what people must do in all circumstances that might arise.
An alternative approach focuses instead on specifying the
constraints within which people have to work (see Panel,
Challenging the traditional approach).
25 Wilson J.R. & Corlett E.N. (Eds) (1995) Evaluation of
Human Work: A Practical Ergonomics Methodology,
Eds, Taylor & Francis
26 Woodson W.E. Tillman B. & Tillman P. (1992) Human
Factors Design Handbook. 2nd Ed. McGraw-Hill
Understanding Human Factors/June 08
Challenging the traditional approach
There are two fundamentally different approaches to
task analysis. The traditional approach, eg HTA, is to
describe tasks in terms of instructions that state what
actions must be carried out in response to stipulated
conditions. This instruction-based approach works best
with what are called closed systems – systems isolated
from the environment. Here, analysts are free to focus
on influences that are internal to the system itself.
The problem is that many systems are open rather than
closed. In other words, they are likely to be affected
by unpredictable events outside the system. The more
open a system is, the more difficult it is to analyse tasks
in terms of instructions that will meet every eventuality.
This is because every eventuality cannot be known. The
impossibility of predicting every event has led some
analysts to switch to constraint-based task analyses.
Instead of trying to write exhaustive instructions
aimed at reducing human error, they try to set out the
operational constraints within which operators can
work safely and flexibly.
To get a better idea of what this means, consider the
difference between navigational directions and maps.
Clear directions allow people to navigate their way
between two points very efficiently, with minimum
mental effort. Maps require people to think more about
what they are doing, but offer several different ways of
completing a journey. Maps are essential in coping with
unexpected events – such as an accident that blocks
the route – while directions become useless.
If you are interested in knowing more about constraint-based methods of
task analysis, the approach is fully described in Vicente (1999).
Page 47
Part 2: Guidance
Design
How is task analysis done?
Stages of HTA
There are three main stages in conducting HTA.
1 Data collection stage – in which information about
the system of interest is collated. A variety of data
collection methods can be used, including observation
(direct or via audio/videotaping), questioning
techniques (informal discussion through structured
interviews), and workshop-based techniques (involving
role-play and walk-throughs). If the system is a
new one, then the data will need to come from a
functional analysis of the system’s goals, supported by
data from pre-existing tasks that are similar.
2 Description stage – in which the results of the data
collection stage are organised into clear statements
describing the task requirements and goals.
3 Analysis stage – in which the task descriptions are
re-expressed in terms of the behaviours that are
required from the user. At this analysis stage, there are
a number of steps that must be carried out (see Panel,
Key steps for a Hierarchical Task Analysis).
What can task analysis be used for?
You can use task analysis to support the following
activities.
• System design/evaluation – in which you can use task
analysis after function allocation (page 37) to help you
understand the human workload (page 125), task
responsibilities and workflow implications of design
decisions. This analysis also shows you where further
design solutions, such as job/task aids, are needed.
Page 48
Key steps for a Hierarchical Task Analysis
Hierarchical Task Analysis involves breaking down the task into a hierarchy of goals and plans.
Stage 1: Identify the goal of the task as a whole
After collating relevant task information (eg via interview or your own experience), you specify the overall task goal,
eg ‘Repair faulty track circuit’.
Stage 2: Identify the sub-goals and plan for the task as a whole
You can now break down the top-level task goal into a sequence of sub-goals (four or five is a convenient number),
together with a description (a plan) of how they should be carried out. Sub-goals for the repair task identified above
might be:
1 Check track circuit status
2 Diagnose source of problem
3 Contact signaller to agree repair
4 Carry out repair and test
5 Contact signaller to complete repair
Stage 3: Break down sub-goals
You break down the sub-goals into even lower level sub-goals until you reach an appropriate level at which to stop.
This level is usually self-evident and is simply the level which yields the most detail needed for the purposes of the
analysis. At each level the sub-goal is also a sub-task, which is also the goal for a set of even lower-level tasks.
Step 4: Describe plans
Once you have set out the full structure of the sub-goals, you need to specify plans for how they are to be achieved.
At its simplest, a plan might be ‘Do each step in order’. A more complex plan for the example at Stage 2 above, might
be ‘Do Steps 1 to 5 in sequence. If the test at Step 4 fails, go back to Step 2’.
• Training design/evaluation – in which you can use task
analysis to generate training scenarios, content, skill
and knowledge requirements, and criteria. If the reason
for the task analysis is as part of training needs analysis
(page 55), you will want to focus on tasks that are the
most likely to be performed incorrectly and have the
highest error cost.
• Interface/workplace design/ evaluation – in which you
can use task analysis to help you decide what demands
(sensing, associating, interpreting, remembering and
responding) user-equipment interfaces (page 33) and
workplaces (page 41) impose on your target users.
For interface design, the analysis might concentrate
Understanding Human Factors/June 08
Part 2: Guidance
Design
on tasks that are mission-critical, difficult to perform,
complex or novel.
• Job design (page 50) – in which you can use task
analysis to help you decide which tasks you can assign
to individual jobs without risking workload (page 125)
that is too high or too low. Task analysis can also
support your decisions about team design.
• Personnel selection (page 79) – in which you can use
task analysis to provide information about the mental
and physical demands of tasks. This can then be used
to identify appropriate criteria against which you can
compare the applicants.
• Supervision and appraisal (page 68) – in which you can
use task analysis to develop scenarios for assessment
of particular skills and knowledge and to provide a
rationale for performance audits.
• Human reliability analysis – in which you can use task
analysis to provide human error data for each task
component. This can then be used to help predict
the future reliability of the system. For hazardous
operations (HAZOPS) and training needs analysis
(page 55), the tasks associated with the severest health
and safety risks to personnel might be selected.
Non-human factors specialists can successfully tackle
most parts of a task analysis: it is more about applying
a logical thought process than understanding human
factors data. However, you may need the help of human
factors professionals in working out the implications of
the analysis for training methods, selection testing, and
workload issues involved in job and interface design.
Stammers & Shepherd (1995) say that task analysis
Understanding Human Factors/June 08
creates the following kinds of information:
• Identification of tasks and their sub-task components
• Grouping of components – an organised, often
hierarchical, listing of the sub-tasks involved in a task,
showing how sub-tasks cluster around goals and over
time, and which sub-tasks share common methods
• Importance, priorities and criticality of sub-tasks
• Frequency of sub-tasks – possibly in different
conditions, such as degraded working
• Sequencing of sub-tasks – either serially or via
conditional branching
•
•
•
•
•
• Information generated by each sub-task – the items of
information that users must supply and its sources
• Knowledge employed in making decisions – the
information that users must use in decision making,
both from displayed sources and from memory
• Knowledge of the system employed in performing subtasks – the understanding that the users must have of
the system and its functions in order to fulfil their role.
Stammers & Shepherd (1995), in Wilson & Corlett (1995), reproduced with
permission
What is Cognitive Task Analysis?
Hierarchical Task Analysis (HTA) focuses on what
people can be seen to do and documents their actions.
Decisions that must be made – eg for taking one
However, when tasks are complex, it is not enough to
branch rather than another
simply observe what people do. It is also important to
Trigger conditions for sub-tasks – a new sequence may find out how they think and what they know, how they
organise and structure information, and what they seek
need to start as a result of another finishing, or as a
to understand better. This is where what is known as
result of a new environmental event
Cognitive Task Analysis (CTA) comes
Goals for each task/sub-task –
in.
‘Sometimes, it is not
goals are established for each
task/sub-task, resulting in a
enough to simply observe The aim of CTA is to describe
hierarchy of goals in increasing
both the cognitive – or conscious
what people do … this
levels of detail
thought – processes that underlie the
is
when
you
might
need
performance of tasks and the thinking
Performance criteria for each
skills needed to respond adeptly to
Cognitive Task Analysis.’
sub-task – criteria need to be
complex situations. The advantage
specified for each goal so that it
of this is that designers get a much
is clear when it has been achieved
better insight into the way people think about their tasks
and what information they need to carry them out. This
Information required by each sub-task – the items of
can lead to much better interface layout designs and
information needed, and their sources, eg displays
more useful job aids.
Page 49
Part 2: Guidance
Design
One of the more well known methods of CTA is ACTA
(Applied Cognitive Task Analysis) developed by Klein
Associates. ACTA is an interview process that is used
with expert users (sometimes known as subject matter
experts – SMEs) to find out about the mental demands
of tasks in order to inform new designs.
Tools and techniques
See Part 3 for more information on data collection
techniques for focus groups, interviews, workshops,
observational analysis and questionnaires. Part 3 contains
more information on task analysis techniques including:
Hierarchical Task Analysis (HTA), Applied Cognitive Task
Analysis (ACTA), Team Cognitive Analysis and Link Analysis
Further information about task analysis
1 Diaper D. & Stanton N.A. (2003) The Handbook
of Task Analysis for Human Computer Interaction.
Lawrence Erlbaum Assoc Inc
2 Kirwan B. & Ainsworth L.K. (1992) A Guide to Task
Analysis, Taylor & Francis, London
3 Shepherd A. (2002) Hierarchical Task Analysis. Taylor &
Francis, London
4 Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
5 Vicente K.J. (1999) Cognitive Work Analysis: Toward
Safe, Productive, and Healthy Computer-Based Work.
Lawrence Erlbaum Associates
Page 50
Job design
What is job design?
Job design specifies the tasks and roles that form an
individual’s job. Its purpose is to ensure that jobs are
appropriately structured and achievable.
Job design often follows function allocation (page 37) (in
which decisions are made about what is to be performed
by machines and what by humans) and task analysis
(page 47) (in which task steps, sequences and criteria
are specified). An essential step in function allocation is
role definition, in which related tasks are collated into a
set of duties that can be assigned to jobs. Role definition
is particularly important when existing jobs are radically
changed by automation or the introduction of new
management structures.
Good job design involving human factors analyses will
pay huge organisation dividends in terms of effective and
efficient operations (see Panel on page 51, A new railway
job). Conversely, failing to consider job design can have a
serious effect on morale (page 117), staff retention (page
87) and safety.
What should be taken into account?
The main considerations in creating effective job design
involve several of the other aspects of human factors
discussed elsewhere in this Guide. These aspects are
brought together in the following list of questions.
• How can targets be set and measured? Are they fair
and achievable? What supervision and appraisal (page
68) procedures will be used to support this?
• Do the tasks that make up each role over-burden
any individual in the system? Too much workload
(page 125) will cause physical and/or mental fatigue,
degrade operator performance, and generally reduce
workplace system efficiency. You particularly need
to be aware of the possibility of overburdening in
emergencies – and make sure adequate experienced
judgement and simulation are used to avoid it.
• Is the work made harder by the way it is organised?
For example, are any required shift work patterns (page
128) designed to enable people to perform at their
best? Are rest and meal breaks adequate? Are the
tasks arranged so as to minimise boredom? People can
become just as fatigued by boredom as by overwork.
• Is the work organised to provide reasonable
opportunity for the individuals involved to experience
some form and degree of self-fulfilment? Job rotation,
job enlargement and job enrichment are examples of
work organisation strategies. These strategies attempt
to ensure that personnel are allocated duties that
provide sufficient variety, autonomy, complexity and
responsibility to maintain motivation (page 117) and
performance. Jobs also need to be structured in a way
that is compatible with available career development.
• Have all the training needs been identified and
properly addressed with adequate training plans and
provision? Training needs analysis (page 55) should start
with the task analysis output and identify the skills and
knowledge that will require training, together with
recommendations on the training content.
Understanding Human Factors/June 08
Part 2: Guidance
Design
Further information about job design
1 Kroemer K.H.E. & Grandjean E. (1997) Fitting the Task to the Human (5th Ed) Taylor Francis
A new railway job
The basic design of many railway jobs (eg signaller, train
driver) has remained fundamentally unchanged for many
years. Sometimes, however, a re-organisation can lead to
a job with new characteristics. Some years ago, as part of
an upgrade to the signalling system, it was proposed to
close the old mechanical signal box at Barnes and move
the functions performed there to the Wimbledon ASC.
The old signal box involved two jobs: the signaller and a
crossing keeper. Between them, they were responsible
for an unusual sequence of five CCTV controlled level
crossings. The crossings were supposed to be operated
in sequence, but their closeness meant that at busy times
this was impossible to achieve. When the functions of the
Barnes box were moved to Wimbledon, it was hoped
that these two jobs could be combined into a single job.
It was argued that the improvement in the technology to
be used by the signaller would mean that this could be
achieved without unacceptable risk.
Several scenarios were developed and used as the basis
for walk-through analysis with a panel of experienced
signalling staff, including those from Barnes. To assist
in these walk-throughs a simple cardboard mock-up
of the facilities that would be used by the signaller at
Barnes was developed. The participants were asked
On the basis of all the information gathered, the second
to act out the actions required by the signaller as the
phase – the human reliability analysis – was carried out. The scenario unfolded and to identify the points at which
implications of each change (two- to one-man operation, they thought the signaller would be unable to cope. They
shift to NX panel, etc) were assessed for all safety critical
were also asked to develop working strategies (eg how
tasks in terms of their impact on workload and the
and when the crossings were operated relative to train
risk of error. The error risk assessment was based on a
movements) to identify the most efficient methods of
human error checklist identifying all perceptual, control,
operating the panel.
communication, decision making and event recording
errors.
The conclusion of this assessment was that the risk of
the signaller being overloaded was within acceptable
The initial risk analysis identified that the mental workload
limits provided a number of conditions were met. The
(page 125) placed on the signaller was the primary risk
three most important conditions were that (a) the panel
area. To assess this more fully, the Simplifier was analysed
was designed so that a second person from the ASC
could assist the signaller in their duties when peak traffic
Before this proposal could be acted on, it was necessary to identify the periods of peak traffic to be regulated.
This was expressed in terms of both train movements
times coincided with other problems adding significantly
to carry out an assessment of the safety implications of
to their workload; (b) the signallers should be trained
this change in job design. This assessment was performed and crossing operations. Discussion with signal box staff
in the most effective working strategies to adopt; and
by two human factors specialists. The first phase involved had identified factors which occurred fairly regularly and
(c) the timetable should be developed to avoid the
gathering all relevant details of the current and proposed which increased workload. These included such factors
as pedestrians and vehicles attempting to ‘beat the
particular combinations of train movements that caused
working arrangements. The old working situation at
barriers’, bridge strikes, barrier failures, T2 possessions, and the highest levels of workload for the signaller.
Barnes and the new situation at Wimbledon were
analysed and compared. The task at Barnes was analysed additional traffic due to diversions. Problems involving the
using Hierarchical Task Analysis. This analysis was based
junction outside Barnes station were judged to impose the
heaviest additional workload.
on observation and debriefings of the existing staff,
supported by training documentation.
Understanding Human Factors/June 08
Discussions with the project team identified the critical
changes to the work that would result from the move to
Wimbledon, such as the move to NX panel working, the
provision of more integrated communications facilities and
the introduction of ‘one-touch’ crossing controls.
Page 51
understanding
Training
human factors
understanding
Training
human factors
Part 2: Guidance
Training
Training
Training needs analysis
The diagram below focuses on Training. It shows three
sub-areas (in the middle red ring), and identifies the main
human factors questions that this Guide answers (in the
outer grey ring). In this Guide we take a broad view of
training. In particular, we view supervision and appraisal
as an opportunity to facilitate continuous training rather
than simply a management activity. At the end of each
section, you will find sources of further information.
What is training needs analysis?
Focus on training
Training Needs Analysis (TNA) helps you to identify
training requirements and their implications for everyone
who works in your organisation. It is a flexible procedure,
with a choice of supporting tools and techniques (see
end of this section). You can use it repeatedly (eg to
support different stages of a user-centred design (page 25)
or staff appraisal process (page 69), generating clearly
defined products and providing an audit trail for all
training decisions.
The diagram shows three key stages and
products.
Operational task analysis produces a list
of tasks, together with how well they
need to be done (ie the required
performance standards).
Training gap analysis produces
a set of training objectives
whose achievement will close
any gap between the existing
performance and the new required
performance.
Training options analysis examines
alternative ways of accomplishing the
training, and finds the option that is
most cost-effective.
These stages are explained in a little more
detail in the section on how to do a TNA.
Understanding Human Factors/June 08
Stages of training needs analysis
Do you need to do it?
Have a look at the checklist in the Panel on page 56,
Do you need to do a TNA?. If you can answer yes to all of
these questions, you almost certainly don’t need to dwell
on this section of the Guide.
When is training needs analysis done?
TNA may take place for two main reasons:
1 To identify training needs as part of the staff appraisal
process (page 69). Here the aim is to ensure either
that any skills or knowledge gaps in staff performance
are addressed, or that staff are developed for future
roles in the organisation as part of their career
development. In either case, the training needs analysis
should proceed in conjunction with the organisation’s
staff appraisal process.
2 To identify training needs due to the planned
acquisition of new equipment or processes. In this
case, the training needs analysis should proceed in
conjunction with a user-centred design process (page
25) for the new system.
Page 55
Part 2: Guidance
Training
Do you need to do a TNA?
TNA is a wide-ranging process that helps you to
identify training requirements and their implications
for everyone who works in your organisation. If you
can answer ‘no’ to any of these questions, you should
consider carrying one out.
• Is there a clear process for analysing the training
needs of all staff at every level in your organisation?
• Are training needs analysed at the organisation, team
and individual levels? If they are, can you show that
they have led to measurably improved workplace
performance?
• Does the training needs analysis process have clear
links with the Finance and HR departments?
• Are managers at all levels skilled in analysing training
needs for those for whom they have responsibility?
• Are training needs prioritised in line with the
achievement of organisational objectives?
• Are the training needs that arise from new initiatives,
projects and equipment all analysed for all their
stakeholders?
• Are appropriate targets set for the length of time
taken between the identification of training need
and its final delivery?
• Does your organisation have a system whereby all
training and development needs are recorded, and
updated once the need has been addressed?
Adapted from HM Inspectorate of Constabulary (1999), reproduced with
permission. © Crown copyright material is reproduced with the permission of
the Controller of HMSO and the Queen’s Printer for Scotland
Page 56
‘Here we outline two
methods for doing TNA ...
a standard method and a
quick and easy method.’
How is training needs analysis done?
Here we outline two methods for doing TNA. The first
method is a standard approach that can be used for
either of the purposes above (ie staff appraisal or to
support new system design). The second method is a
quick and easy TNA process that takes only a few hours
(with the right participants) and which you will find useful
if you are just interested in reviewing the training needs
of existing job holders (ie in support of staff appraisal).
Standard TNA
Standard TNA goes through the three stages shown in
the earlier diagram on page 55. For new system design it
is normal to carry out Operational Task Analysis several
times as more detailed prototypes are developed and
more information becomes available, before attempting
the other two stages. However, it is important that the
required training is developed in conjunction with the
plans for any new system roll-out.
Stage 1: Operational Task Analysis
Operational Task Analysis should produce:
1 A list of tasks involved. If more than one job is affected
(eg if there is to be a major new system), you should
produce a list for each job affected. Each task should
carry a description of the operational circumstances
under which it is performed, together with the
Using subject matter experts
To obtain information on required knowledge, skills
and abilities, you can interview job supervisors, HR
managers or, when appropriate, existing experienced
operators. Often the best procedure is to supply a
panel of five or six knowledgeable people with a list of
the tasks, and ask the following type of questions:
• What does a person need to know in order to do
the task?
• What do you expect a person to learn in training
that would make them effective at this task?
• What are the characteristics of good and poor
operators on this task? A useful way of getting at
this kind of information is via the repertory grid
technique – see Further information).
• Think of someone you know who is better than
anyone else at this task. Why are they able to do it
better?
operational standard to which it must be performed.
It may be helpful to use one of the methods of task
analysis (page 47) for this stage. Where appropriate,
you may wish to use a panel of subject matter experts
(see Panel, Using subject matter experts).
2 The interface and workplace assumptions of the
new operational requirements in terms of physical,
functional and environmental factors (you may need
human factors specialists for this). The assumptions will
need to be identified in conjunction with the systems
engineers as part of a user-centred design process
(page 25).
Understanding Human Factors/June 08
Part 2: Guidance
Training
3 The training priority for each task based upon its
Difficulty, Importance and Frequency (often called a
DIF analysis). You will need to work with one or more
experienced operators in producing this DIF analysis.
Stage 2: Training Gap Analysis
This produces Training Objectives which should specify
the additional training requirements you will need to
address, compared with current arrangements. These new
training requirements should be described in terms of the
knowledge, skills and attitudes that people will need to
acquire.
Stage 3: Training Options Analysis
This produces a recommended training solution from
a range of possible options. You should initially choose
options on the basis of an analysis of the training
objectives and the constraints of current organisational
training policy. For example, if distance learning is the
preferred training policy, then the media will have to be
compatible with this method of training. If, however, it
could be beneficial to amend policy, then this should be
open for discussion. The options analysis should identify
the most cost-effective option. It does this by comparing
the Training Effectiveness and Cost-Benefits of each
option (see below).
The output from the options analysis is a recommended
training solution. This should incorporate the following:
• Option specification – a description of the media
options that will partially or fully meet the training
requirements as described by the training objectives
(see Panel, Selecting training media).
Understanding Human Factors/June 08
Selecting training media
Selection of training media boils down to a choice
between three broad approaches:
• Telling the trainee what to do using verbal methods, eg
lectures, discussions, notes
• Showing the trainee what to do by demonstration or
guidance, eg animation, video
• Having the trainee practise what to do, eg using role
play, simulators, part-task trainers.
The strengths and weaknesses of different technological
solutions to these approaches are shown in the Panel
on page 58.
• Training Effectiveness Analysis – an evaluation predicting
the relative training effectiveness of each option based
on comparisons against similar proven media systems,
and/or research carried out with new technologies. You
can use prototyping, experiments, research of literature
and comparisons with existing training to predict or
assess performance. This is, by its nature, a subjective
process, which makes it important to consult
appropriate training analysts and subject matter
experts for the best advice. You will need to consider
all the criteria that contribute to an ideal solution and
then rank how well each option meets every criterion.
Later sections in this Guide on training methods and
training delivery mechanisms will help here.
of confidence or tolerance. The costs should then
be offset against any benefits, such as the potential
for income generation. In the absence of any actual
numbers, you can do a Relative Cost-Benefit Analysis
that simply compares the options with each other in
terms of High, Medium and Low costs and benefits.
(You’ll find a method for doing this in the Teamworking
Journey Guide published by RSSB and available from
their website.)
• Selection of the most cost-effective option – a clear
description of the advantages and disadvantages
of each option, with supporting arguments for the
recommended solution. You need to examine the
trade-off between training effectiveness and costbenefits, and then decide on the optimum solution in
terms of cost-effectiveness. See the section on costeffective training (page 61) for comprehensive guidance
on this aspect of training.
• Cost-Benefit Analysis – a list of the major costs for each
option, covering capital, staffing, infrastructure and
support costs over the projected life of the training
system. Each cost should have a risk factor, ie a level
Page 57
Part 2: Guidance
Training
Strengths and weakness of different training technology options
Technology option
Description
Strengths
Weaknesses
(CBT)
Training delivered via personal computer or
(web) server. Materials are presented as a
mix of text, graphics, photographs, animation,
video and sound tracks, and pre-sequenced
for the learner by the author(s). Often
includes some form of branching to provide
different sequences dependent on what the
trainee has learned so far.
•Improved training standards (quality, consistency, trainee attention and retention)
•High up-front development costs (hardware, authoring
system, software design, programming, editing, testing).
These costs increase in line with the complexity of the
media
•Standalone systems
•Networked systems
•Complex task procedures presented very effectively via video and animation
•Efficient use of courseware (eg multiple language commentaries)
•Reduced training costs (timeliness, availability, higher throughput, reduction in central facilities
and personnel, 30% faster training, reduced failure rates)
•Easier training management (trainee monitoring, courseware revision, course validation)
•Risk-free learning environments
•Learning on a computer in your own time and in comfortable settings can be very effective
•Trainees may have more flexibility in using their own learning style and can select learning
materials that meet their level of knowledge and interest
•Trainees can study anywhere they have access to a computer and Internet connection
•Trainees can test out or skim over materials already mastered and concentrate efforts in
mastering areas containing new information and/or skills
•Learning objective tests can be incorporated in the teaching software
•Release of instructor time for effective tasking: debriefing, reviewing, discussing and directing
•Costs escalate if materials need frequent revision
•Trainees require certain level of computer literacy
•Trainees cannot ask questions or brainstorm ideas
•Lack of human/trainer contact, which can greatly affect
learning for some
•Trainees with low motivation or bad study habits may fall
behind
•Slow Internet connections or older computers may make
accessing course materials frustrating
•Managing computer files and online learning software can
sometimes seem complex for trainees with beginner-level
computer skills
•Portable and flexible
Intelligent CBT
Simulators
•Part-task simulators
•High fidelity
simulators
Embedded
training
•‘Stimulation’
•Manual help systems
•Automatic and semiautomatic contextsensitive help systems
Page 58
Training delivered by personal computer
or (web) server. Includes some form of
exploration environment for the trainee
to explore the material, ‘watched’ by the
computer. Sequence of materials decided
by the computer on the basis of observed
trainee performance.
•As for CBT
•As for CBT
•Optimised learning progress (materials customised for individual trainees ‘on the fly’)
•Less mature than CBT – very restricted choice of system
Dedicated hardware/software that simulates
a dynamic task environment with near
perfect fidelity. Part-task simulators simulate
selected parts or dimensions of the task with
near perfect fidelity (ie non-critical parts of
the task are represented only symbolically).
•As for CBT
Software that uses the real task equipment
to present help or training materials relating
to user tasks.
Automatic context-sensitive help systems
anticipate user requirements and intervene
with suggestions or explanations, triggered by
user error or delay.
•Training time can increase (although quality of learning is
better)
•Safety critical/rare procedures can be safely practised
•High fidelity simulators are very expensive to develop
and build and maintain
•Increases utilisation of operational equipment (ie real equipment not used for training)
•It may be difficult to ask questions or discuss ideas
•Very high transfer of training to real tasks
•Full simulation is intimidating for non-experts
•As for Simulators (in the case of embedded simulation, sometimes known as ‘stimulation’
since fake data are used to stimulate real displays)
•User has to take effective control of own learning
•Saves time and increases productivity (readily accessible, relevant help whenever required)
•Reduces or eliminates need for paper reference manuals
•User needs to know what they are trying to achieve for
help to make sense (automatic help systems)
•User needs to understand enough of system to know
what help to ask for (manual help systems)
Understanding Human Factors/June 08
Part 2: Guidance
Training
Strengths and weakness of different training technology options
Classroom-based
courses, seminars
and workshops
Classroom-based courses, seminars and
workshops are generally held off-site
(without normal office distractions and can
typically last from a couple of hours to a
whole week or more)
•Classrooms can be enjoyable and comfortable settings in which participants can get better
acquainted with their company and their fellow employees. This leads to greater readiness for
training
•Low cost
•Can ask questions and brainstorm ideas
•Trainees can develop analysis and decision-making skills, learn from mistakes and discover
principles and concepts for themselves
•Trainees cannot use own pace or sequence
•Dedicated sessions can be used for (self) assessment,
though this is usually not appropriate in a workshop or
seminar
•Classroom-based processes may not suit individual
learning styles.
•Some trainees find classroom environments stressful
•Trainee tracking is not very practicable
•Increased retention rates
•Classroom-based training cannot prepare trainees for the
miscalculations, delays, and other obstacles they inevitably
will encounter on the job
Quick and easy TNA
Quick and easy TNA was developed by Via International
Ltd and is reproduced here by permission. It takes just
a couple of hours or so and is especially useful for
reviewing the training needs of existing job holders. This
means it is best suited to the staff appraisal process. It is
less suited to supporting new system design, although
it could be useful for the appraisal of new system
prototypes by experienced holders of related jobs (eg
drivers appraising a new cab design).
‘Quick and easy TNA
takes just a few hours
and is especially useful
for reviewing the training
needs of existing job
holders.’
• Post-it notes for each participant
Process
The process comprises four steps:
Step 1: Generate issues (30 minutes)
Ask each participant to list the issues they face in their
work. Issues should be written on post-its, one issue per
post-it.
Illustrate what type of response you expect, eg lack
of understanding about how to perform a particular
procedure, lack of information about the network,
failure of management to respond quickly to queries,
inappropriately targeted marketing (give one training and
one non-training related issue as an example).
The process avoids asking participants directly what
training they want or need. Instead it focuses on issues
that may be addressed by training (emphasising that
training is but one solution to the issues faced by the
organisation). Here’s how you do a quick and easy TNA.
• Coloured dots (three colours, 48 dots per colour
– one sheet – per participant to start with; make sure
you have spare sheets to hand)
• Index cards (one per participant) for forced ranking
Confirm that instructions are clear.
Preparation
• Markers (at least three colours) to write on flip charts
Allow 15 minutes (including set-up instructions). In the
meantime, you can put flip charts on the walls.
You will need:
• Prepared flip charts (about 10 pages, 4 columns
on each flip chart page: issues, training, non-training,
combination)
Understanding Human Factors/June 08
• A group of participants sufficiently experienced to
generate informed insights about their tasks.
Ask participants to post their notes on flip charts in the
first column (allow 5 minutes).
Page 59
Part 2: Guidance
Training
Go through the issues, reading each one out and asking
participants to take off their issue if it is similar; maintain
physical movement of participants, (allow 10 minutes).
Note 1: Avoid discussion at this stage. Any issue is valid, even if only
one participant thinks so!
Note 2: Number the notes in sequence - but don’t try to group
anything yet.
Step 2: Identify training/non-training solutions (30 minutes)
Explain that the next step is to decide for each issue
whether it can be addressed by:
• Training – when an issue is caused by a lack of
knowledge, skills, ie I don’t know how to... (red dots)
• Non-training – when an issue is caused by inappropriate
policies, inadequate communications, inappropriate
remuneration, lack of clarity in responsibilities, etc.
(blue dots)
• A combination of training and non-training (white dots).
Ask participants to assess for each issue whether it can
be addressed by training (red), non-training solutions
(blue), or a combination of the two (white).
Step 3: Clarify results (30 minutes)
Explain that there is no time to discuss each issue
and assessment in detail, but that you need high-level
agreement on the outcome of the previous steps, ie ‘Are
there any assessments (or issues) that you absolutely
disagree with (and why)?’
Note 4: Look for issues where there is a wide spread of different
assessments. Avoid discussion on areas where people already mostly
agree that training, or a combination of training and non-training, should
be used.
Note 5: Identify issues that should be totally or partly addressed by
training (circle the letters identifying these issues), and add any that get
resolved as such via this part of the discussion.
Emphasise that the focus for today is to identify potential
areas for training, other issues will be documented and
fed back, but cannot be discussed in detail within the
scope of this workshop. However, recognise the issues as
important to the participants!
Step 4: Force-ranking of training needs (15 minutes)
Hand out index cards (one per participant).
Note 3: Don’t use red and green dots together because colour-blind
people will not be able to distinguish between the two.
Ask each participant to review the issues that have been
identified as having full or part training solutions (ie the
circled ones) and list the five issues that they feel should
be addressed within the next six months. They should
prioritise these five issues by allocating a value of 5 for
the most important and 1 for the least important.
Ask people to come up to the flip charts and stick the
dots on the post-it notes to reflect their assessment for
each issue (allow 15 minutes).
Collect the cards and explain that you will review the
results and combine certain areas where similar training
can address multiple issues.
Hand out sheets with coloured dots for each participant.
Tools and techniques
• Relative Cost-Benefits Analysis - described in the
Teamworking reference (later in this Guide)
• Hierarchical Task Analysis
Further information on training needs analysis
1 Jankowicz D. (2003) The easy guide to repertory grids,
Wiley
2 Gregory D. & Shanahan P. (2004) Teamworking best
practice in the railway industry: The Journey Guide,
Gregory Harland Ltd, for RSSB, Euston
3 Patrick J. (1992) Training Research & Practice,
Academic Press Ltd, London
4 RSSB (2007) Good Practice in Training: a guide to the
analysis, design, delivery and management of training,
RS/220 Issue 2, June 2007
5 RSSB (2007) Good Practice Guide on simulation as a
tool for training and assessment, RS/501 Issue 2, June
2007
6 Via International Ltd (unpublished) developed the
‘quick and easy’ training needs analysis described in
this section. Via International, Building 3, Chiswick Park,
566 Chiswick High Rd, Chiswick, London W4 5YA
End.
Page 60
Understanding Human Factors/June 08
Part 2: Guidance
Training
Cost-effective training
What is training?
Most definitions agree that training is the development
of knowledge, skills and attitudes required to perform a
specific task or job to a specific standard.
The goals of training are quite different from those of
education. Training is concerned with getting an individual
to home in on a specific performance in a specific
context. Education is concerned with getting an individual
to open up to new possibilities that are essentially
unspecified in both content and context. Training
succeeds by eradicating individual differences; education
seeks to identify and exploit them (where they are usually
referred to as ‘talents’). Any confusion between training
and education is due to the fact that they both share the
requirement that people are able to learn.
How do people learn?
It is tempting – even usual – to imagine that people
learn through the transmission of ‘knowledge’ from some
source (eg a book, a computer screen, a teacher) into
their brains. In fact, learning depends on people’s ability to
develop meaning for the patterns of events, activities and
relationships all around them. This is a complex, individual
process which trainers and teachers can either help or
hinder. People learn because they can create meaning and
integrate it with what they already know, not because
they simply receive information.
Furthermore, it appears that what is learned is
fundamentally connected to the conditions in which it
is learned. Like a bird with a nest, learning is built out
Understanding Human Factors/June 08
‘Recently, I was asked if I was going to
fire an employee who made a mistake
that cost the company $600,000. ‘No,’ I
replied, I just spent $600,000 training him.
Why would I want somebody to hire his
experience?’ Thomas J. Watson, Founder of IBM
of the materials to hand and constructed to reflect
the constraints of local conditions. For humans, these
constraints include the organisational rules and working
conditions imposed upon them, and the way their
colleagues respond to these rules and conditions in
working practice. People learn not just what to do and
think, but how to do and think those things ‘around here’
so as to be acceptable to their managers and workmates.
(See the sections on Why do people break rules? on page
12), and Culture on page 91 and Conditions on page 117).
A lot of research has revealed which factors are the most
important for effective learning (see Panel, Key factors for
effective learning).
Key factors for effective learning
• Doing – to rehearse our developing skills and knowledge
• Failing – with feedback to give us insight into the progress and nature of our learning
• Reasoning – to allow us to generate explanations for failure
• Trying again – to allow us to test and improve our explanations for failure
• Well-told stories – to allow us to connect together what we learn and embed it in our everyday knowledge
• Just-in-time instruction – to prevent our newly acquired skills and knowledge from fading
• Progressive fidelity – to give us the right amount of realism to support our learning progress. Too much realism is
a waste of money and overwhelms novice learners; too little realism prevents people fine-tuning their expertise,
and may encourage them to depend on responses that do not transfer very well to the operational task
environment. They may end up depending on cues that never occur in reality
• Access to resources – to provide us with demonstration, explanation or further detail whenever we might benefit
from them
• Emotional predisposition – to ensure that we are motivated and energised to engage with the learning process
• Sympathetic context – to ensure that external conditions and media for learning are suited to us (eg for many,
sitting at a VDU or in a busy office is not a sympathetic context for some stages of learning)
Adapted from Naughton, reported in Ison (2002), reproduced with permission
Page 61
Part 2: Guidance
Training
How should people be trained?
How should you design training?
There are three aspects to the design of an effective
training programme:
Guidance on training method components
Training components
Design guidance
Define the pre-conditions
for training
•Be clear about the capabilities and limitations of the specific trainee audience.
1 Proper understanding of the trainees (see How do
people learn? on page 61)
•Do not limit your attention simply to the entry criteria for the start of the training: make sure that you have
defined the criteria for transition to each of the stages of the training.
2 A relevant and well structured training content (see
Training Needs Analysis on page 55)
3 The application of an effective training method (ie
ways of organising and presenting training materials
so that we maintain trainees’ attention and maximise
their learning progress).
Unfortunately – and despite a great deal of research –
there is no agreed best training method. While we can be
clear about what components a successful method must
Is part-task better than whole-task training?
On the basis of common sense you would expect
part-task training to be better than whole-task training.
However, what this overlooks is the subsequent
problem of how to help the trainee put together the
various parts.
Where there is high interdependence between
different parts of the task, it may be better to arrange
for the trainee to work with successive approximations
of the whole task. You then introduce more nuance and
detail across the whole task as performance improves.
Part-task training is a good bet when performance on
different parts of a complex task have little effect on
each other.
Page 62
•Take steps to understand their interests, motivations, expectancies, hopes and fears about the training. Ensure that
training materials are attention-getting and communicate the training objectives and trainee benefits clearly and
appropriately. Doing these things will help to ensure that trainees are ready for the training.
•For knowledge-based training, consider the use of ‘advance organisers’ before each new stage of training. These
are organising concepts that give previews of what is to come and help trainees develop a way of integrating the
detail of what they are to learn under training. In developing advance organisers, try to use concepts that the
trainees already understand.
Design and present the
structure and sequence of
the training materials
The objective here is to structure and sequence a body of knowledge or a skill so that the trainee can progressively
grasp it. You will need to use your understanding of the subject matter to decide how to simplify the information
to be learned and organise it into chunks that will be manageable for the trainees. For training involving a new
software system or software application, you may wish to consider an alternative approach to training design based
on the minimalist approach to training (see Panel on page 63, The minimalist approach to training).
Gagné’s skill acquisition hierarchy
Skills training
Skills are often acquired via:
•A thinking phase, where the required performance is discussed. You can support the thinking phase by clear
descriptions and explanations of the principles, objectives and techniques of the required performance.
•A doing phase, where task performance, with feedback, becomes increasingly rapid and error free. For tasks that
can be separated into discrete stages, you can support the doing phase via part-task training (see Panel, Is partbetter than whole-task training?). Here, the skill is broken down into parts that are practised separately before the
well-rehearsed parts are then put together. Several variations are possible. In progressive part-task training, the
first and second parts are practised separately before being combined and practised together. A third part is
then practised separately before being added to parts one and two – and so on. Cumulative part-task training is
the same as progressive except that there is no pure-part training. So part one is practised, then parts one and
two, then parts one two and three etc. Finally, retrospective part-task training starts with practising the last part to
which is then added the penultimate part and so on. Retrospective part-task training can be useful for motivating
some people since they are quickly rewarded with successful completion of the task
•A tuning phase, in which normal performance becomes automatic and less demanding of conscious thought,
and thus able to cope with more unusual events. You can support the tuning phase by injecting increasingly
subtle interruptions into the task, eg to simulate degraded working. Another good technique is some form of
overtraining in which training is continued past the point where the performance criteria are met. It sounds as
though ‘overtraining’ is wasteful, but it is not. This is because it increases resistance to stress, fatigue and other
interference during subsequent performance of a task. Using a simulator, you can achieve overtraining by ‘above
realtime’ training in which the task or scenario is run faster than in real life. This can be very effective in producing
stress resistance.
Cont./
Understanding Human Factors/June 08
Part 2: Guidance
Training
Guidance on training method components
/Cont.
Knowledge training
You may find it helpful to structure and sequence training materials according to a hierarchy of knowledge
elements, such as that worked out by Gagné (see diagram in this table, also Further information).
Essential features to be picked out from their surroundings (discriminated), such as alarm sounds or visual cues,
may be isolated or exaggerated to make them more obvious initially. Materials to support the learning of concepts
and rules require the use of a variety of examples that emphasise generalisations and exceptions. Rules should be
applied to a variety of examples to ensure that the trainee understands each rule and its use.
You should ensure that the structure used can support exploration and navigation by the trainees.
Provide appropriate levels
and types of feedback
Facilitate retention and
transfer
Feedback (sometimes known as knowledge of results) is an essential part of training. Without it there can be no
learning. There are five different types of feedback which you should be concerned with:
Trainee question
Feedback type
Trainer guidance
What is that (for)?
Display
Explain objects and their functions
What should I be looking at?
Event
Explain changing significance of objects
What should I be doing?
Task
Describe rules, give prompts
What should I have done?
Review
Highlight & review decisions & errors
How well did I do?
Results
Encourage, score and grade
Several factors influence how well skills and knowledge are retained after training. Some enhance retention
while others make skills resistant to decay (often called skill fade). Unfortunately, it’s not clear what the relative
importance of these factors is. But enough research has been done to indicate their general influence (see How
can skill fade be prevented?).
have, there are almost as many ways of implementing
those components as there are training designers.
The Panel on these pages, Guidance on training method
components, shows what components a training method
must have, and gives some practical guidance on how to
address these.
Note, however, that you should not interpret the
components in the Panel as serial steps, but as areas for
attention. For example, you would normally give feedback
in the course of presenting the training materials. You
would also take opportunities for retention and transfer
at every stage of training design and its execution.
Understanding Human Factors/June 08
How should you design training for different people?
People differ in their capacity to learn new things. This
capacity may be natural and related to ability. However, it
can also be related to experience – as when, for example,
people have bad memories of school – or to age. Older
people are often slower in acquiring new skills, and
previously learned materials may interfere more with new
learning.
In the Panel, Guidance on training design for trainees with
different needs, you’ll find information on how to adapt
training design to better suit people who for one reason
or another are less able to learn new things.
The minimalist approach to training
During the 1980s, a group at the IBM Watson Research
Center developed a ‘minimalist’ approach to training
design. They realised that the natural problem-solving
behaviour of people could be used to support the
goals and activities of users rather than trying to teach
them everything there was to know about the tasks
that could be accomplished by the software. They used
their analysis of this problem solving to identify five
minimalist principles, as follows.
1 Make sure learners can get started quickly. Reduce repetition; eliminate
non-essential instructions; offer the learner meaningful activities as
soon as possible.
2 Rely on learners to think and improvise. Encourage learner inference;
leave out material that can be inferred; don’t give the learner
understanding – allow them to create it.
3 Embed information in real tasks. Introduce real work immediately;
recognise that people will bring real goals to the training situation
– so allow them to achieve them.
4 Take advantage of what learners already know. Make connections
between new concepts and those already understood from previous
software systems; ensure that full use is made of concepts already
covered in the current training (sometimes known as ‘scaffolding’).
5 Support error recognition and recovery. Mistakes cannot be avoided, but
they can confuse and frustrate; if harnessed they can be very useful.
Creating minimalist instruction involves a task analysis
to understand what learners are trying to achieve,
how they are likely to approach their tasks and what
mistakes they are likely to make. Minimalist training
design entails building a sequence of real world tasks
with minimal documentation. This means that it involves
users from the beginning in the very tasks they need to
accomplish with the software.
Summarised from research by Carroll (1990)
Page 63
Part 2: Guidance
Training
Guidance on training design for trainees with differing needs
Potential trainee difficulties
Guidance on training design
Tasks depend on short-term memory
•Avoid verbal learning and the need for conscious
memorising. Use cues which guide the trainee.
•Avoid part-task training. If it’s essential, use cumulative parttask training.
•Ensure enough testing has taken place before moving on to
the new material.
Tasks suffer from interference from other
training activities or prior learning
•Restrict the range of activities in the training course.
•Employ longer learning sessions (ie longer periods without
interruption).
•Provide variety by changing the teaching method rather than
the subject matter. Change in content may lead to confusion.
Information must be translated from one
medium to another (eg from the training to
the operational environment)
•Ensure visual materials are consistent with operational reality.
Learning is abstract or unrelated to realities
•Present new knowledge only as a solution to a problem that
is already appreciated.
Tasks are ‘paced’
•Allow trainees to proceed at their own pace.
How can skill fade be prevented?
Completing a training course is one thing. But people tend to forget what they have
learned. Even experienced operators will perform less well on aspects of the task that do
not crop up very often. And some skills and knowledge seem to fade faster than others if
not practised. The research on ‘skill fade’ is by no means complete: you cannot look up any
table that lists all the knowledge and skills and tells you how often they must be rehearsed
to keep them fresh. However, there is quite a bit of useful guidance.
Factors affecting skill fade
Skill fade factors
Training guidance
Training factors
Related to the way the task
is formally taught and to
differences in teaching methods
•The more coherent (ie connected) the content of a learning or training
programme, the less likely it is that the knowledge will be forgotten.
•If you use simulation, make sure that it is sufficiently realistic
to enable the trainee to relate what is learned to the
operational task.
•Grouping pieces of information into categories increases the amount of
information recalled.
•Short sessions (1 hour max per day) can induce better skill acquisition than long
or multiple sessions.
•The more that the training content can be elaborated (ie with additional
information) or rehearsed during training, the more durable trainees’ recall of it is
likely to be. But there is a trade-off between this and using the available training
time for new material – it needs an experienced trainer to judge this
•Allow trainees to structure their own programme if possible.
•Encourage trainees to beat their own targets rather than
those of others.
•For complex tasks, explanations of how things work result in better recall of task
procedures after training than simply procedural training on its own. Simple or
well-rehearsed procedural tasks do not benefit from functional explanations.
Tasks get more complex
•Allow trainees to learn by easy stages of increasing
complexity.
•Training on different pieces of similar equipment decreases the rate of acquisition
but increases retention.
Trainees lack confidence
•Use written instructions.
•Feedback about the direction and scale of error (eg ‘too much’, ‘too little’, rather
than ‘right’ or ‘wrong’) improves trainee performance over subsequent trials.
•Avoid using operational materials too soon.
•Giving feedback in summary form rather than after every practice trial slows
performance but promotes a marked improvement in retention.
•Use longer induction periods.
•Assign groups of workmates to the same course.
•Gradually fading out the amount of feedback, instead of providing either a
constant or increasing amount of feedback, improves retention.
•Avoid formal tests.
•Don’t give formal time limits for course completion.
Learning requires extended attention and
concentration
Adapted from Newsham (1969). © Crown
copyright material is reproduced with the
permission of the Controller of HMSO and
Queen’s Printer for Scotland.
Page 64
•Avoid ‘chalk & talk’ as far as possible. In earlier years trainees
may have had a bad experience of the classroom.
•Use a learner-centred, discovery learning style – even in the
classroom setting.
•Use meaningful material and tasks which are challenging to
an adult.
Skill maintenance factors
Related to activities undertaken
by the individual after the task
is learned
•Using skills on the job immediately after training prevents skill fade.
•Mental rehearsal throughout training enhances subsequent task performance. It’s
best to combine it with physical practice and it should not be allowed to exceed
five minutes.
•Having access to embedded training facilities means that people always have the
means to practise aspects of required skills, including rarely occurring events.
/Cont.
Understanding Human Factors/June 08
Part 2: Guidance
Training
Factors affecting skill fade /cont.
Skill fade factors
Training guidance
Re-training factors
Related to the way the task is
formally re-learned/re-taught
and to differences in re-learning/
re-teaching methods
•The more experienced a re-trainee is, the better their performance after retraining.
Individual factors
Related to characteristics of the
individuals learning the tasks
and differences between these
individuals
•High aptitude people learn more quickly than low aptitude people. They learn more within the same period of time
and retain more for longer.
•Expecting and receiving an evaluation test increases retention.
•Retraining on the same task will take much less time than first training (often about half).
•Psychomotor tasks (eg driving) are re-learned faster than procedural tasks
•People with similar levels of knowledge but different levels of aptitude forget the same amount of information within
the same period of time.
•Adding noise and variation during practice improves retention for people who are impulsive rather than reflective
– probably because it forces impulsive people to be more reflective than they otherwise would have been.
•Retention is highest when the operational conditions (eg stress, noise etc) under which people must recall information
are the same as those employed during training.
Task factors
Related to the task
characteristics and differences
between tasks
•Retention is improved if high-quality job aids are provided.
•The fewer the number of steps in the task, the better the retention.
•The less rigid the sequence of steps in the task, the better the retention.
•Retention is improved if each step in the task has built-in feedback.
•Retention is improved if the task is not time-limited.
•The less the mental processing involved in the task, the better the retention.
The table on these pages, Factors affecting skill fade, lists
the types of factor that affect skill fade, and gives guidance
on how to control their influence.
How should training be evaluated?
Evaluation of training plays a critical role in enabling your
organisation to spend large sums of money effectively, so
helping it to assure its own future. Just as human learning
depends on knowledge of results to develop effective
individual performance, so your organisation requires
feedback on its training investments in order to develop
effective organisational performance.
Ideally, the evaluation of training should proceed at
four levels (see Panel on page 66, Four levels of training
evaluation). However, many organisations fall short of
carrying out training-related evaluation beyond the first
level. Some fail to carry out any evaluation at all, though
there can be understandable reasons for this.
•The less the difficulty and number of facts and terms to be remembered, the better the retention.
•The less the motor control demands of the task, the better the retention.
Context factors
Related to the context in which
the task is learned, performed
or recalled. This context can
be internal (ie mood state),
external (ie environmental) or
task-related (ie contextual task
demands such as time)
•Training designers need to establish a context that is similar to the operational one in which trainees will be expected
to recall the content of the training programme.
Summarised from research by Gregory Harland Ltd (1999), © Gregory Harland Ltd, reproduced with permission
Sometimes, organisations fail to carry out proper
evaluations of their training investment due to practical
difficulties. For example, the training may be administered
by a contractor who does not have any access to the
trainees once they return to the workplace. Or the time
lag may be too long between the training and the posttraining assessment to be of any use in deciding how to
modify the training. Or there may not be sufficient staff,
resources or operational opportunity to carry out the
job assessments.
In other cases, there may be organisational difficulties.
For example, there may be resistance to the collection
of field data due to a fear of liability among management
Understanding Human Factors/June 08
Page 65
Part 2: Guidance
Training
Four levels of training evaluation
Evaluation level
Description
Implementation
Trainee reaction
This measures the degree of satisfaction with the
course and its presentation. It is a useful indicator
of course popularity and the quality of the
presenter and materials used on the course.
Typically measured by a course feedback questionnaire.
Learning objectives
This is sometimes called internal validation.
It measures the ability of the training course
to achieve the learning objectives set for it. It
is possible for a training course to get a very
positive trainee reaction (they all had a good
time), but very low internal validation (they didn’t
learn what they were supposed to).
Internal validation uses learning objectives derived from a training needs
analysis, possibly itself informed by performance criteria from a task analysis. It
can be carried out via three different kinds of test:
•A pre-test (or assessment process) establishes the pre-course performance
of the trainee
•A post-test establishes the difference the course made to the achievement
of the learning objectives
•A retention test establishes the level of performance against the learning
objectives some time later. Note, however, that retention or ‘skill fade’
factors can be influenced by the design of the training in the first place
– see earlier in this section.
Job behaviour
This is sometimes called external validation.
It measures the ability of the training course
to deliver the required levels of operational
performance. It is possible for a training course
to have high internal validation (trainees learned
exactly what they were supposed to) but low
external validation (what they learned is of little
benefit in the workplace).
Typically measured via a staff supervision and appraisal process (page 68)
– which has its own sets of problems.
Organisation
function
This is the highest level of training evaluation.
The impact of training is measured in terms of
organisational criteria such as production time
and quality, damage to equipment, absenteeism,
staff turnover, safety statistics, and staff attitudes. It
is possible for a course to have high internal and
external validation (it does what is says and is of
measurable value in the workplace) but scores
badly at an organisational level (eg operators
get overstressed and frequently go sick, or
infrequently used equipment gets damaged
through skill fade).
Typically measured via corporate statistics and staff surveys. You should
compile the data in a way that establishes its relationship with the existence
(or absence) of specific training interventions. For example, absentee data
should include information that allows managers to deduce any trainingrelated causes.
Page 66
for what might be discovered. Or there may be financial
pressures that prevent management trying to find out
if people need retraining. It is sometimes tempting to
decide that what you don’t know will probably hurt you
less than what you might find out.
An obvious alternative to post-training evaluation is to try
to do it while the training is still underway – but there are
some difficulties with this. For example, a trainee doing
well in a performance test on the training course may
not be reflected in their operational performance. This
is often because the benefits of recent intensive practice
are lost with time.
Another difficulty is that training courses are often
designed to keep performance levels artificially high, by
limiting the type and number of mistakes that trainees can
make – for reasons of motivation (page 117)and safety.
This can lead both trainees and trainers to overestimate
the true levels of trainee expertise (see Panel, The
Longford Explosion). However, with care, the quality of
in-training test data can be improved in order to estimate
post-training effectiveness. You can do this in two main
ways:
• You can give trainees greater insight into their own
learning. This will help to prevent them confusing
their feelings of familiarity with the ability to recall
information. Through feedback from regular in-course
testing, it will also help them to understand the need
for, and value of, making mistakes, allowing them to
get a sense of the impact of the training on their own
performance. (Actually, in-course testing that demands
recall rather than recognition – as in multiple choice
testing – increases the value of training, because
Understanding Human Factors/June 08
Part 2: Guidance
Training
The Longford explosion
In 1998, the Esso Gas Plant exploded in Longford,
Australia.The immediate cause of the explosion was
a ruptured heat exchanger which had become very
brittle due to intense cold. The cold had been caused
by an earlier problem in the oil circulation system.
The results of the inquiry showed that operators did
not understand what could happen when metal got
cold - even though they had been through what at first
appeared to be a proper training course.
Closer inspection by the inquiry revealed that if
operators failed the test at the end of the training
module they were given further coaching. Importantly,
though, operators were re-assessed simply by asking
them if they now understood the training. In the event,
operators reported they felt under some pressure to
say that they did.
The inquiry questioned the operators further and
discovered that they had not grasped the fundamental
meaning of what they needed to know. For example,
operators knew enough to say that they took specific
actions to prevent ‘thermal damage’ but it turned out
they could not explain the concept of thermal damage.
They were simply remembering what they had been
told rather than what it meant.
The basic problem was that the assessment was testing
the wrong thing – rote memory, rather than a deep
understanding of the job.
Understanding Human Factors/June 08
recalling material from memory makes such recall
more likely in the future.)
• You can also educate trainers about the use of intraining testing. However, particular care is needed
for tests of skill (rather than knowledge), since such
tests can be very unreliable predictors of long-term
performance. Training designs that use closer and closer
approximations to the whole operational task, and are
practised and tested under variable conditions, offer
much better predictions of long-term performance
(and therefore post-training evaluation) than part-task
training regimes.
Further information about cost-effective training
6 As of May 2008, http://tip.psychology.org/theories.html
Summarises 50 or so different theories of learning
developed by a large number of psychologists and
researchers in the field of human learning
7 Newsham D.B. (1969) The Challenge of Change to
the Adult Trainee. Training Information. Paper 3. HMSO
8 Patrick J. (1992) Training Research & Practice,
Academic Press Ltd, London
9 Seely Brown J. & Duguid J. (1991) Organizational
Learning and Communities of Practice: Towards a
Unified View of Working, Learning, and Innovation, The
Institute of Management Sciences (now INFORMS)
1 Air Affairs Ltd (2003) Safety critical roles: refresher
training and re-assessment frequency, RSSB Research
report, Rserv479. Available at www.rssb.co.uk/pdf/
reports/research/safety%20critical%20roles%20-%20
refresher%20training%20and%20re-assessment%20
frequency.pdf (as of May 2008)
2 Buckley R. & Caple J. (1992) The Theory and Practice
of Training, Kogan Page
3 Carroll, J.M. (1990)The Nurnberg Funnel. Designing
Minimalist Instruction for Practical Computer Skill.
Cambridge, MA: The MIT Press
4 Gagné R.M. Briggs L.J. & Wager W.W. (1992) Principles
of Instructional Design. Holt, Rinehart & Winston
5 Gregory Harland Ltd, (1999) Development of a
skill fade model, GHL/CHS/SkillFade/Deliverables/
FinalReport/Volume1/v3.0, Centre for Human
Sciences, DERA
Page 67
Part 2: Guidance
Training
Supervision and appraisal
• building effective teams
• contributing to the formation of organisational policy
We have placed supervision and appraisal in this
section on training to emphasis their crucial role in the
continuous development of front line staff in the railway
industry. Several key supervisory skills are covered in this
section of the Guide. Together, they equip a supervisor
to facilitate the training development of all the staff for
which they are responsible.
• fostering a relationship which allows both supervisor
and staff to learn from mistakes via constructive
feedback.
• consulting and briefing staff on wider organisational
information and safety issues
Support includes:
• assisting staff through the complaints process.
What should a supervisor actually do?
At first sight, supervision seems to be all about
monitoring and controlling the work of others. In fact,
Morrison (1993) says that to be an effective supervisor
you need four distinct skill sets:
• Management (see page 99)
• Staff development
• Support
• Mediation.
Management includes:
• managing the time and workload of staff
• assessing their competence by reference to
professional, organisational and legal requirements
• ensuring that records are completed.
Staff development includes:
• appraising the performance of staff to identify their
training needs
• dealing properly with complaints about staff
• creating a safe climate for staff members to reflect on
their working practices and their effects on self and
others
Underpinning all of these four functions is the essential
skill of clear and effective communication (page 106).
You’ll find further information and guidance on various
• conducting ‘difficult conversations’ and helping them
topics that need to be understood by supervisors
to talk about their feelings,
elsewhere in this Guide. As well
‘Supervision is about
especially fear, anger, sadness or
as communication, these topics
helplessness
include training needs analysis
managing staff, developing
(page 55), teamworking (page
• recognising and supporting
them, supporting them, and 103), stress (page 120), and
staff who may suffer from
workload (page 125).
mediating on their behalf ’
abuse – whether physical or
psychological
This section provides human factors good practice
• monitoring overall health and emotional behaviour
of staff, especially with regard to the effects of stress
(page 120)
guidance in areas that are particularly relevant to
supervision – namely:
• advising when staff might need to seek external
counselling.
• Performance appraisal and teambuilding (used in staff
development)
Mediation includes:
• Difficult conversations (used in support)
• briefing more senior management about resource
problems
• Negotiating skills (used in mediation).
• Competence assessment (used in management)
• linking staff with other parts of the organisation,
including higher management
• negotiating and developing team purpose and scope
Page 68
Understanding Human Factors/June 08
Part 2: Guidance
Training
What’s the difference between competence
assessment and performance appraisal?
As a supervisor, you may also be guided by the use of
validated task descriptions via a task analysis (page 47),
skill and knowledge inventories, verbal test questions, and
The difference between these two is really all a matter
guidelines on (for example) the
of what the purpose is.
number of times a person needs
‘Performance appraisal looks
Competence assessment is
to correctly perform a task to be
concerned with regulation: it is at past performance to specify
deemed competent.
a management concern with
current development needs;
establishing that an employee
RSSB maintains several good
has the knowledge, skills and
competence assessment looks practice documents on
attitudes necessary to perform
assessment and appraisal - see
work to the standard expected. at current behaviour to assure
Further information.
In the railway industry a code
future performance.’
of practice exists (ORR, 2007)
for competence assessment.
What should competence assessment cover?
Performance appraisal is concerned with staff
development. Its purpose is to ensure that people are
motivated in their jobs and are developed to their full
potential.
Competence assessment should cover:
Most organisations have well-defined procedures for
both competence assessment and performance appraisal.
Some of the methods underlying these procedures
overlap (eg in the use of rating scales).
• underlying knowledge, such as understanding the
relationship between interacting rules
However, competence assessment relies more on direct
observation of the operator at work (either on the job
or in a simulation) to assure future performance, while
performance appraisal relies more on interview and
discussion that draws on records of past observation to
identify development needs. Both procedures are usually
based upon pro-formas and a pre-defined schedule for
their use. The procedures often rely on being heavily
informed by the supervisor’s own experience in the job.
Understanding Human Factors/June 08
• skills, such as being able to demonstrate an ability to
(say) interpret display readings, diagnose faults, operate
controls, enact a procedure
• The ability to carry out a prescribed procedure of
work can, usually, be demonstrated by a ‘show me’ test,
in which people attempt to complete the task.
• Mental skills, such as the ability to assimilate signal
control information from an IECC and then interpret
it, might be demonstrated by the operator talking
through the interpretation of displayed information.
However, this may interfere with some mental skills.
It may also not be possible to verbalise other mental
skills, such as mental arithmetic. In these cases posttask debriefing of operators may be appropriate.
• Completion of a task requiring knowledge, such as
fault diagnosis, may be indicative of underpinning
knowledge. However, it is possible that the correct
action was by luck. Accordingly, knowledge should be
assessed through verbal or written questioning.
• Psychometric personality tests may provide a
prediction of interpersonal, team management and
safety behaviours. However, observing actual behaviour
in the real or simulated work setting using behavioural
observation tends to provide a more valid measure.
• safety behaviours and attitudes, with regard to (for
example) communication and teamwork.
Greenstreet Berman (2003), reproduced with permission. © Crown copyright
material is reproduced with the permission of the Controller of HMSO and
Queen’s Printer for Scotland.
How should competence assessment be carried out?
What should performance appraisal cover?
Greenstreet Berman (2003) say that appropriate
competence assessment methods depend on whether
the focus is on knowledge, skills or attitudes.
• Physical skill competencies can be demonstrated by
practical ‘show me’ assessments, in which people either
complete the real task or a piece of it, such as setting a
route on an NX panel or troubleshooting a signal.
To carry out performance appraisal you should:
• discuss how the individual’s job objectives contribute
to organisational goals
• discuss the organisation’s expectations regarding
employee performance
• provide feedback to the individual about their job
Page 69
Part 2: Guidance
Training
performance in light of organisational objectives
• coach the individual on how to achieve their job
objectives or meet the requirements of their job
• diagnose the individual’s strengths and weaknesses
• determine what kind of development activities might
help the individual to make better use of their skills
to improve performance on the current job – and in
preparation for future jobs.
How should performance appraisal be carried out?
Hansen says you should carry out performance appraisal
via the following four inter-related steps:
1 Establish a common understanding with the employee
regarding work expectations – mainly, the work to be
accomplished and how that work is to be evaluated.
2 Assess performance and the progress against work
expectations on an ongoing basis. Provide regular
feedback to clarify and modify goals and expectations,
correct unacceptable performance before it is too
late, and reward superior performance with proper
praise and recognition.
3 Produce formal documentation of performance
through the completion of a performance and
development appraisal pro-forma appropriate to the
job.
4 Hold a formal performance and development
appraisal discussion, based on the completed appraisal
form and ending in the construction of a development
plan.
Hansen (undated), reproduced with permission
Page 70
What problems are there in assessment/appraisal?
Two general problems in carrying out assessment/
appraisal procedures arise from:
• insufficient structure
• taking things at face value.
Insufficient structure may be due to an organisation
failing to invest in the task analysis (page 47) needed
to define the critical tasks, together with the criteria
for their successful performance. It may also be due to
the failure of the supervisor to be sufficiently trained
in understanding and using this structure. Structure can
be greatly enhanced by the use of rating scales. Best
practice here is to ensure that the scales are developed
for aspects of performance identified as most important
by the task analysis. Even so, organisations will need to
be aware of a series of biases and other factors that can
make their use less reliable. These are listed later in this
section.
Taking things at face value refers to the supervisor failing
to marshal the necessary evidence for their judgements
about the performance they are assessing. It is never
enough to allow people – either trainees or trainers – to
make face-value judgements about their competence
levels (see Panel on page 67, The Longford Explosion). This
may be because of time pressures, lack of training, or
one or another observational biases to which all human
judgements are vulnerable. All appraisal and assessments
inherently depend on subjective judgements. This means
they are subject to certain problems and biases that you
need to guard against if the results are to be considered
reliable (see Panel on page 70, Appraiser biases).
Appraiser biases
Halo/Horns Effect The halo effect is the tendency to rate
someone high against all criteria because they are high on one or two.
The resulting appraisal is not helpful to employees, since it does not
identify development needs. The opposite effect also happens when
consistently low judgements in one or two categories are allowed to
affect judgement in other areas. This is sometimes referred to as the
‘devil’ or ‘horns’ effect.
Matthew Effect This is related to the Halo Effect and refers to the
tendency of people to judge others over time in the same way they
have always judged them. This tends to create a self-fulfilling prophecy,
with more weight being given to the previous judgement than to any
evidence to the contrary. This is called the Matthew Effect after the New
Testament statement: ‘To him who has shall be given, and he shall have
abundance: but from him who does not have, even that which he has
shall be taken away.’
Interpretation Bias This refers to the fact that different people use
different meanings for terms like fair, good, and excellent. It is best not to
use these terms in any case because they do not help managers to know
what to do about the results.
Central Tendency Effect This refers to the tendency for raters to
assess most people as average. Rating people average can seem safer to
many supervisors. It means they don’t have to have difficult conversations
(page 73) with their staff and it may give them some illusion of control
over very good people. It also means that they can avoid explaining to
their own management why some of their team are poor performers.
Recency Bias This is a tendency to assess people on what they have
done most recently and ignoring behaviour that is less recent..
Leniency Bias This is a tendency to rate people higher than they
deserve. This may be because the supervisor wishes to avoid confronting
people with their limitations, or because they have allowed friendship
to become confused with management, or because they are concerned
that negative feedback will de-motivate their staff.
False Attribution This is the tendency to attribute success or failure
exclusively to what people do and ignore the context and constraints
under which they are working. So if someone does well, we give them
credit, and when someone poorly we say they are to blame. In both
cases it is easy to ignore the circumstances of their performance.
In making the mistake of assuming that good and bad performance
are both under the complete control of the employee we miss the
opportunity of taking a systemic view and of recognising real success or
truly helping someone to improve.
Understanding Human Factors/June 08
Part 2: Guidance
Training
Countering the effects of appraisal bias
There is a series of simple steps you can take to limit the
effects of appraiser bias.
• Be aware of the problem. The first line of defence lies
in raising awareness of the problem. All staff – not
just supervisors – should be informed of the types
of subtle bias that can interfere with performance
judgements. If everyone involved understands the
existence and influence of biases on human judgement,
much can be done to overcome or challenge them.
• Use better rating scales. Rating scales that grade people
from poor to good should be avoided. Far more
preferable are forced choice rating scales that make
supervisors select a pre-defined statement that is
nearest to the observed performance. Even better
are behaviourally-anchored rating scales. These mark
points on the scale with actual job behaviour relevant
to the job-holder being rated. The job behaviours used
on the scales are usually developed from observations
of a range of people on the job. The supervisor who
uses such a scale can ‘anchor’ their assessments of each
individual in an unambiguous, more objective way that
means more to both parties.
• Get evidence for judgements. All assessments should
be backed up with objective evidence for supervisor
judgements. In competence assessments, this evidence
needs to be drawn from the behaviour observed
on the job (or its simulation) during the assessment
session. In performance appraisal, it may be helpful
to use the critical incidents technique. Here, critical
performance categories are identified, and during the
evaluation period the supervisor records examples of
Understanding Human Factors/June 08
critical behaviours in each of the categories. This log is
then used in an evidence-based discussion to evaluate
the employee at the end of the evaluation period.
• Use multiple sources. Better assessments and appraisals
can be made by combining multiple sources of
information. For example, evidence-based peer reviews
are usually highly acceptable to all participants and
fairly accurate as well as relevant to the task in hand.
Evidence-based self-reports using structured forms
can work well for people with a clear understanding of
their own work. Placing self and supervisor judgements
alongside each other can be a very useful basis for
discussion.
‘360 degree feedback’ is a comprehensive but
expensive type of appraisal. It includes ratings by both
self, peers, and subordinates, as well as the more
traditional, downward assessments from line managers.
It gives people a chance to know how they are seen
by others and to become aware of their skills and
style. It may also improve communications (page 106)
between people. However, it only works for people
with at least three people reporting to them and may
be uneconomic for all but senior managers.
• Incentivise the development of poor performers. The
organisation may offer incentives (financial or nonfinancial) to encourage supervisors to make special
efforts to help poor performers improve.
How do you build an effective team?
Guidance on how an organisation can diagnose and
improve teamworking is given elsewhere in this Guide
(page 103). Of particular relevance in this section
on supervision and appraisal is a particular aspect of
teamworking – teambuilding. Any search of the Internet
will produce a large number of sites offering information
in this area. From a human factors point of view, the most
useful guidance is on the general process that a team
needs to go through, which can be considerably helped
by effective supervision.
How do teams develop?
Teams seem to go through a number of stages as their
members develop together. The Panel on page 72, Stages
in team development, describes these stages and how
supervisors can best deal with the typical things that go
on in their teams at these different stages. What is meant
by ‘team’ here is a group of people who work together
over a relatively long period (weeks, months or longer),
such as a track gang, station staff, a signal box shift, an
administrative team, or staff put together for a special
project. The development process described here is
especially relevant to those teams for which there is any
kind of opportunity for interpretation of the Rule Book
or other organisational procedures.
Page 71
Part 2: Guidance
Training
Stages in team development
Stage
Description
Likely supervisor feelings
Effective supervisor behaviour
Forming
When a new team is first formed, people’s own agendas dominate. They may
not declare all of their interests. The supervisor’s key role here is to get team
members oriented – towards one another and their joint responsibility (team
task). The key focus should be on this team task, time scales and resources.
Progress on the team task may seem slow.
•Anxiety about being liked
•Clarify and explain task objectives and demands
•Anxiety about the abilities of the
team members.
•Make people feel included – identify and use previous experience of new team members
•Create ‘getting to know you’ opportunities
•Clarify and explain communication channels – often a problem area for new teams
•Be directive but don’t be dictatorial: create opportunities for participation
•Take the initiative and allow the team to depend on you
•Relax – avoid seeming over-anxious
Storming
Norming
Performing
Deforming
Reforming
Where there is room for manoeuvre outside the Rule Book, people will
disagree about the best way to carry out the team task. Supervisors can find this
frustrating and difficult to manage. But conflict has its benefits – the emotions
often result in greater team cohesiveness. The supervisor should focus on
ensuring that participants understand the team’s terms of reference. Supervisors
should also expect their own performance to be challenged.
•Anxiety about where the team is
going and how it will get there.
•Allow differences of opinion – but do not allow extreme views to develop.
•Anxiety about how much control
should be exerted.
•Respond to feedback seriously, and insist that it is constructive – distinguish between
yourself as a team members and your role as supervisor
•Anger, frustration – maybe apathy.
•Accept responsibility for your role as supervisor
The clash of personal agendas gives way to a more collective team concern with
how they are going to work together. Each team member starts to accept that
the needs of the whole group are larger than their personal issues. Team focus
turns to the timing, pattern and quality of operational matters. Some storming
will continue for remaining issues. Supervisor attention should be focused on the
achievement of the required quality standards.
•Relief, sense of getting somewhere
at last.
•Allow the team to develop norms rather than rules: don’t let behaviour get rigid
The team has replaced its many personal agendas with a creative focus on team
problems via mature procedures. Team members will have developed trust that
allows them to review progress frankly. They can effectively deal with setbacks and
devise remedial steps. The team will feel at home with each other and enjoy their
success. Leadership often becomes a shared function with different members
taking the lead for those task elements that best fit their skill sets..
•Contentment, involvement,
commitment, relaxation – but some
anxiety about taking more of a ‘back
seat’.
•Challenge existing norms – allow the team to consider degraded/emergency operations
Teams deform for a number of reasons. The project may have finished or a
key member leaves. Teams develop their own ways of marking their successes
and changes. If publicised, these can make the team attractive to outsiders
and a desirable one to join. If someone leaves due to promotion, this can be
confirmation that the team is a desirable place to be. Trying to carry on as normal
may cause problems, and the supervisor will need to pay attention to these.
•Celebration, success, pride – and
anxiety about ‘having to start again’.
•Recognise the change as a positive endorsement of the whole team
Reforming is necessary when the team changes in membership or scope.
Reforming is similar to the forming phase but is often easier due to familiarity.
Supervisors may need to correct a common assumption by team members that
the new person is simply filling in for the person who left, when in fact they have
their own contribution to make. An older team facing a new task may be overconfident. It may be necessary to take the team back to its previous decisions to
allow it to storm its way again to a new set of norms that will pave the way to
further success.
•Concern over how the change will
affect team performance.
•Respond to challenges constructively – harness the expertise of the whole team
•Continue to focus attention on assuring a constructive feedback climate for everyone
•Use the performance appraisal process to reinforce team roles
•Let team members begin to ‘have their head’ – but check results against standards
•Don’t let the team get too focused on itself – develop relationships with others
•Allow others to lead and let go as far as possible – but check the results!
•Do what you can to create more sense of equality between team members
•Involve the team in discussing the implications for the forthcoming change
•Monitor team performance during the change and discuss problems as they emerge
•Involve the team fully in the discovery and resolution of these problems
•Interest in making other changes that
will benefit the team or organisation.
•Anxiety and frustration about having
to start again – albeit with much
more confidence than before.
•Involve the team in full discussion of the change. If a new team task, work with the team
in understanding its challenge and impact on existing work patterns and practice. If a new
member, make opportunities for personal contact with the rest of the team.
•Take the initiative in valuing the new member’s ideas, contribution and perspective.
•Encourage the team to spot opportunities for using the change to resolve existing
problems and make improvements.
Summarised from work by Tuckman (1965) and Morgan et al (1994)
Page 72
Understanding Human Factors/June 08
Part 2: Guidance
Training
How do you handle a difficult conversation?
What is a difficult conversation?
Stone et al (1990) say that a difficult conversation is
anything you find hard to talk about. The context might
be reviewing bad performance with someone, disciplining
someone, having to make someone redundant who is
also your friend, dealing with the bereaved spouse of an
accident victim who was your colleague and so on. But
whatever the context, the conversation will be difficult
whenever you feel vulnerable or it has implications for
the way you feel about yourself, when the issues are
important but there is uncertainty about how things
will turn out, and when you care about the people with
whom you need to have the discussion.
Research has shown, maybe surprisingly, that despite
the many variations, all difficult conversations share a
common structure. Understanding this structure makes it
possible to conduct such conversations more effectively –
and with much greater confidence. In fact, in each difficult
conversation, Stone et al (1990) say there are really three
different conversations going on. In each of these, we all
tend to make predictable errors that distort our thoughts
and feelings, and make the conversation much more
difficult than it need be. This can lead us to far more
unsatisfactory outcomes than necessary.
What are these common threads?
1 The what happened? thread. Here we spend a lot of
time wrestling with each other over who is right, who
meant what and who is to blame. The problem is that
we tend to make incorrect assumptions which cause a
significant amount of the difficulty we experience.
Understanding Human Factors/June 08
2 The feelings thread. Very often we try to suppress the
strong feelings we have within a difficult conversation.
We tend to think this is better this since expressing
them may lead to temper and loss of control, or
else things being said which are then regretted. The
problem is that feelings are at the very core of difficult
conversations. They are not a by-product of engaging
in a difficult topic: they are an integral part of the topic
– and they need to be recognised as such.
3 The identity thread. Our role in the conversation affects
our sense of who we are and how we are perceived
by others. Something beyond the subject of a difficult
conversation is always at stake. And that something
Making difficult conversations easier
is us. For example, our determination to turn down
a proposal from a keen young team member is
undermined when we are suddenly reminded of
ourselves at their age and the way we would have felt.
We are overcome by self-doubt and our anxiety level
suddenly soars. We lose our balance. In mild cases,
our confidence leaves us, we lose concentration and
forget what we were going to say. In extreme cases,
we may feel paralysed, overcome by panic and ‘wish
for the earth to open up and swallow us’.
So what can be done? The Panel, Making difficult
conversations easier gives some good clues.
A battle of messages
A learning conversation
Assumption: I know all I need to know to
understand what happened.
Goal: Persuade them I am right.
Assumption: Each of us is bringing different information and
perceptions to the table; there are likely to be important things
that each of us doesn’t know.
Goal: Explore each other’s stories: how we understand the
situation and why.
Asssumption: I know what they intended.
Goal: Let them know what they did was
wrong.
Assumption: I know what I intended and the impact their actions
had on me. I don’t and can’t know what’ss in their head.
Goal: Share the impact on me and find out what they were
thinking. Also find out what impact I’m having on them.
Assumption: It’s all your fault. (Or it’s all
my fault.)
Goal: Get them to admit blame and take
responsibility for making amends.
Assumption: We have probably both contributed to this mess.
Goal: Understand the contribution system: how our actions
interact to produce this result.
The Feelings Conversation
Challenge: The situation is emotionally
charged
Assumption: Feelings are irrelevant &
wouldn’t be helpful to share. (Or my
feelings are their fault & they need to hear
about them.)
Goal: Avoid talking about feelings. (Or let
them have it!)
Assumption: Feelings are the heart of the situation. Feelings are
usually complex. I may have to dig a bit to understand my feelings.
Goal: Address feelings (mine & theirs) without judgements or
attributions. Acknowledge feelings before problem solving.
The Identity Conversation
Challenge: The situation threatens our
identity
Assumption: I’m competent or
incompetent, good or bad, lovable or
unlovable. There is no in-between.
Goal: Protect my all-or-nothing self image.
Assumption: There may be a lot at stake psychologically for both
of us. Each of us is complex, neither is perfect.
Goal: Understand the identity issues on the line for each of us.
Build a more complex self-image to maintain my balance better.
The What Happened? Conversation
Challenge: The situation is more
complex than either person can see
Reprinted by permission of International Creative Management, Inc. Copyright © 1990 by Stone, Patton, and Heen.
Page 73
Part 2: Guidance
Training
How do you negotiate effectively?
What is negotiation?
Negotiation is a communication between two parties
designed to reach an agreement when both parties
share some interests, but not others. Fisher et al (1991)
say that less obvious than this straightforward idea is the
notion that negotiating does not mean giving in. Nor
does it mean settling for less than you wanted – or even
compromising. However, it does often mean settling
for something different – sometimes more – than your
original objectives. If this is surprising, it is because it is
common to have an incorrect sense of what negotiation
is all about. The usual assumption is that the subject of
negotiation is like a pie, and the object of negotiation is to
get as much of the pie as possible. And, of course, getting
more means that the other side gets less. But in fact, it’s
better to think of negotiation as a creative process in
which opportunities are developed and expanded so that
the interests of both parties are met. In its ideal form, it’s
the process of making the pie big enough for everybody.
The four principles of negotiation
Negotiating principle
How to do it
1. Separate the people from the
problem.
•Put yourself in their shoes – it helps to see through the ‘merits’ of your case and the ‘faults’ on the other side.
It is people who negotiate. Yet
people have strong emotions,
different perceptions and problems
communicating. Emotions often
get mixed up with the merits of a
problem, and taking positions just
makes this worse since personalities
get identified with positions.
•Don’t infer their intentions from your fears – it is too easy to think the worst of the other side and this will
seriously impede progress.
•Don’t blame them for your problem – it may be justified, but it is almost always counterproductive.
•Discuss each other’s perceptions – it may reveal new values and options.
•Look for opportunities to not always act in accordance with the other side’s perceptions – this will dislodge their
tendency to ‘pigeonhole’ you.
•Include the other side in the problem-solving process as early as possible (ie avoid having to face the ‘not invented
here’ syndrome (what’s this?) when you present your ideal solution).
•Make your proposals consistent with their values – help them to avoid having to save face.
•Recognise and understand emotions (yours and theirs) – it will give you valuable insights towards the solution.
•Make emotions explicit and acknowledge them as legitimate – it reinforces each side’s humanity.
•Allow the other side to let off steam, but don’t react to emotional outbursts – ensure only one side at a time gets
angry.
•Use symbolic gestures – an apology can vastly improve a hostile emotional situation.
•Listen actively and acknowledge what is being said – it will show that you take them seriously and help them to
listen to you.
•Speak to be understood – often achieved by severely limiting the number of people in the same room.
•Speak about yourself, not about them – it is difficult for them to challenge how you feel and very easy for them
to argue that you are wrong about them.
•Speak for a purpose – make sure that you have thought through the impact of what you might say beforehand.
What’s involved in negotiation?
The notion that negotiation is like a pie that must be
fought over goes hand in hand with an approach to
negotiating called positional bargaining. The classic example
is the haggle that often takes place when buying a used
car or something from a car boot sale. The approach
depends on each side successively taking – and then
giving up – a series of positions. While this approach tells
each side what they want and can eventually lead to an
acceptable agreement, it is fraught with serious problems.
2. Focus on interests, not positions.
Positions usually obscure what
people want. Finding a compromise
between two different positions
will often not address the interests
that led the two sides to take those
positions in the first place. It is far
better for people to declare what
they want to achieve and where they
want to get to.
•Ask ‘why?’ and ‘why not?’ – it will lead you from their position to their interests.
•Realise that each side has multiple interests – they are trying to solve problems that are more complex than
might at first be apparent.
•Realise that the most powerful interests are basic human needs – like recognition, security, a sense of belonging,
and control over one’s life.
•Acknowledge their interests as part of the problem – people will listen better if they think you have their interests
at heart.
•Put the problem before your solution – if they hear a position first, they will not listen to your problem.
•Look forward with a purpose in mind, not back with only causes to dwell on – or you will just end up point
scoring off each other.
•Be concrete, but flexible – you need to be clear about what will satisfy you, while always being open to new
suggestions that may turn out to be better.
•Be hard on the problem, soft on the people – commit to your interests, but do not attack people or they will
become defensive and closed.
Page 74
Understanding Human Factors/June 08
Part 2: Guidance
Training
The four principles of negotiation
Negotiating principle
How to do it
3. Invent options for mutual gain.
•Brainstorm options, identify the most promising ideas – then figure out ways to improve on them.
Rather than spending time under
pressure trying to reach an
agreement with the other side, it
often pays to set aside some time to
brainstorm a wide range of options
that advance shared interests and
reconcile differing ones.
4. Insist on using objective criteria.
Negotiations can get stuck if one
side simply decides to be stubborn.
This can be countered if both sides
agree ground rules that allow
matters to be deferred to some
third-party standard, such as an
independent expert, market value
or legislation. Discussing such
criteria, rather than what sides are
(un)willing to do, means that neither
needs to give in to the other. Instead,
both defer to a fair solution.
•Consider brainstorming with the other side – harder to do, but potentially valuable. But make it clear that the
brainstorming session is off the record.
•Broaden options – seek variations by thought experiments which try the options out hypothetically – and then
critique the implications.
•Look for mutual gain – by identifying shared interests and seeing if different interests can be dovetailed (Jack Sprat
could eat no fat; his wife could eat no lean …)
•Decide on fair standards – eg what a court would decide or scientific judgement.
•Decide on fair procedures – eg some version of ‘one cuts, the other chooses’.
•Discuss the standards and procedures with the other side – so that agreed, third-party criteria can be used to
determine progress towards an agreed solution.
• Positional bargaining is inefficient. The process often
takes a lot of time – due to starting with an extreme
position in the expectation that you will need to give
it up for something less, but which is still more than
Understanding Human Factors/June 08
Further information about supervision and appraisal
1 Fisher R. & Ury W. (1999) Getting to Yes: Negotiating
an agreement without giving in. Random House
2 Fletcher C. (2004) Appraisal and Feedback: Making
performance review work, Chartered Institute of
Personnel & Development
3 Goldstein I.L. & Ford K. (2001) Training in
Organisations, Wadsworth
Excerpts from GETTING TO YES 21e by Roger Fisher, William Ury and Bruce Patton. Copyright © 1981, 1991 by Roger Fisher and William Ury.
Adapted and reprinted by permission of Houghton Mifflin Company. All rights reserved.
• Positional bargaining is not guaranteed to produce a wise
agreement. In adopting a position and then defending
it, people tend to get locked into that position. The
transaction then takes on a new dimension as people
try to save face – reconciling future action with past
positions, and less and less attention is paid to the
underlying concerns of the parties involved.
There is an alternative to positional bargaining – called
principled negotiation – that has come to be adopted
in many countries and organisations as a much more
effective and practical approach to negotiation. The four
key points of principled negotiation, and the ways in
which they can be addressed, are set out in the Panel, The
four principles of negotiation. You will find further helpful
information in the main source for this table.
your undeclared objective. All sorts of delaying tactics
are also employed to apply pressure – dragging feet,
threatening to walk out etc.
• Positional bargaining endangers relationships. Positional
bargaining is a contest of will. As one side bends to
the will of the other, there is anger and resentment. A
win for one side can lead to long-term or permanent
damage to the relationship, which makes further
agreements less likely.
4 Greenstreet Berman Ltd (2003) Competence
assessment for the hazardous industries. Research
Report 086 Health and Safety Executive
5 Hansen, D. http://iso9k1.home.att.net/pa/performance_
appraisal.html (as of May 2008)
6 HSE (2002) Railway Safety Principles and Guidance
Part 3A, Developing and Maintaining Staff
Competence, HSE Books
7 Morgan B.B. Salas E. & Glickman A.S. (1994) An
analysis of team evolution and maturation. The Journal
of General Psychology, 120(3), pp 277-291
Page 75
Part 2: Guidance
Training
8 Morrison, T, (1993) Staff Supervision in Social Care.
An Action Learning Approach, Harlow, Essex, England,
Longman
9 Office of Rail Regulation (2007), Developing and
maintaining staff competence, 2nd Edition, ISBN
07176 1732 7
10 Quinones M.A. & Ehrenstein A. (1999) Training for a
Rapidly Changing Workplace, American Psychological
Association
11 RSSB (2007) Good Practice Guide on simulation as a
tool for training and assessment, RS/501 Issue 2, June
2007
12 RSSB (2008) Good Practice Guide on Assessment,
RS/701, Issue 2, Aug 2008, RSSB, Euston, London
13 RSSB (2008) Good Practice Guide on Driver Specific
Assessment RS/702 Issue 1 Aug 2008, RSSB, Euston,
London
14 Robertson I.S. (2001) Problem Solving, Psychology
Press
15 Silberman M. (2003) The Consultants Big Book of
Organisational Development Tools’ McGraw-Hill
16 Stone, D, Patton, B. & Heen, S. (1999) Difficult
Conversations, Penguin Books
17 Tuckman B.W. (1965) Developmental Sequence in
Small Groups, Psychological Bulletin, vol. 63, 1965, pp.
384-399
Page 76
Understanding Human Factors/June 08
understanding
© Four by Three 2005
Staffing
human factors
understanding
Staffing
human factors
© Four by Three 2005
Part 2: Guidance
Staffing
Staffing
Selection
Staffing is concerned with the human factors of
recruiting, selecting and retaining the right people in your
organisation. The diagram below focuses on Staffing. It
shows its three sub-areas (in the middle red ring), and
identifies the main human factors questions that this
Guide answers (in the outer grey ring). At the end of
each section, you will find a list of sources of further
information that will provide more detail. In addition, Part
3 gives further detail on key human factors methods that
are mentioned throughout the Guide.
How do you select the right people?
Focus on staffing
The first thing to say is that unless you are already a
Human Resources (HR) professional, you will need to
work with one – as well as occupational experts – to
assure the quality of your selection process.
When most people think of how to select the right
people, they most often focus on the selection interview.
It is certainly true that the selection interview is by far
the most common method used to select people for
employment. However, it is not best practice to rely on
it exclusively. In fact, in its traditional form, the selection
interview is greatly inferior to several other selection
methods described in this section. Furthermore,
whatever selection method – or combination of
methods – is used, it is important to approach
selection as a process that takes place over
time rather than the means by which an
organisation makes a decision about a job
applicant on a particular day.
What is the selection process?
The selection process runs as follows:
• Carry out a job analysis to identify:
- the most difficult, frequent and important
tasks that make up the job
- the knowledge, skills and aptitudes the person
needs to do the job
- which of the knowledge and skills can be trained
and which the applicant will need to have.
Understanding Human Factors/June 08
• You will almost certainly have to provide some
induction training – so that the new people can
get acquainted with your organisation’s culture and
procedures. If you expect to hire experienced people
(eg COSSs, depot fitters), you may need to provide
only minimal job training. In other cases (eg new
signallers, drivers), job training will form a significant
early phase of employment.
• Decide upon the selection methods that will best
assess the knowledge, skills and aptitudes that will be
needed to carry out the tasks that make up the job.
• Attract as many applicants as possible – the more
there are, the better the choice and the more likely
you are to get who you want (see Recruitment on
page 84).
• Apply the selection methods (sifts, tests, interviews
etc) to the job applicants and on the basis of these
assessments, select those most likely to do the job well.
• Make job offers to the selected candidates, conditional
on references, and follow up on references. While you
are waiting for references to be completed, it is a good
idea to contact the successful candidate: if you want to
employ them, it is highly likely that others do too, so
you will need to make sure they remain interested in
your job offer until they can start.
• Evaluate the effectiveness of the selection methods by
seeing how well they predicted the performance of
those selected and by calculating their cost-benefits.
(Of course, these calculations are hampered by the
fact that you can never know how well the people
whom you didn’t select would have done.)
Page 79
Part 2: Guidance
Staffing
What selection methods are available?
A variety of selection methods are available and there is a
great deal of reliable evidence about their value. The main
methods are:
• interviews
• practical tests
• psychometric tests
• biographical data (biodata) questionnaires
• assessment centres.
Interviews
There are two main kinds of selection interview. Most
organisations use a traditional interview approach in
which candidates are asked a series of straightforward
questions about their previous experience, interests,
aims, ambitions and expectations (see How should you
conduct a selection interview? on page 82). However, some
organisations are increasingly using a competency-based
interview approach (sometimes known as behavioural
interviews). Here, a candidate’s previous behaviour is
sampled in order to indicate their future performance.
They may be asked to describe a real situation from their
experience that required for example, problem-solving,
leadership or stress management. The interviewer will
want to know specifically how the candidate handled
these types of situations – in terms of both their
behaviour and attitude (see Panel, More about competency
interviews).
Page 80
More about competency interviews
It is can be quite effective to combine competency or
behavioural questions with more traditional interview
questions. Skilled interviewers commonly do this. The
questions often overlap and are designed to gather
information about the working issues most relevant to
your organisation. But make sure that any competency
questions you ask do not discriminate unfairly against
people who have not had a chance to experience the
situations you are asking about eg younger applicants.
These are examples of interview questions.
•Initiative and follow-through
Give me an example of a situation where you had to overcome major
obstacles to achieve your goals.
•Thinking and problem solving
Tell me about a time when you anticipated potential problems.
Describe the preventive measures you took to avoid a major problem.
•Communication
Describe for me a situation where you persuaded team members to
do things your way. What was the effect?
•Working effectively with others
Tell me about a difficult situation you had with a co-worker, and how
you handled it.
•Leadership
Tell me about a time when you were able to step into a situation, take
charge, muster support and get good results.
•Priority setting
Tell me about a time when you had to pick out the most important
things in some activity and make sure that they got done.
•Decision making
Describe for me a time when you had to make an important decision
on the basis of limited facts.
•Ability to work in varying work conditions (stress, changing deadlines, etc)
Tell me about a time when you worked effectively under pressure.
•Delegation
Describe for me a time when you had to delegate to a person with a
full workload, and how you went about doing it.
Adapted from Reed (2003), reproduced with permission
Practical tests
These are only appropriate for candidates who already
claim to have the skills necessary to do the job. Practical
tests directly assess the abilities required on the job. For
example, a fitter might be asked to undertake a repair
task on a piece of equipment (or simulator), a technician
might be asked to troubleshoot a faulty signal, or a
secretary might be asked to carry out a task using wordprocessing software.
Psychometric tests
Psychometric tests can be useful when choosing a
candidate from a group of people who do not have
easily comparable skills or experience. You normally need
a specially trained and certified person to administer
them, and using them often involves a fee. Psychometric
tests are used to measure intelligence, personality or
aptitude for specific tasks, such as decision making, and
interpersonal skills. A Guidance Note on psychometric
testing for train driver selection and management is
in an advanced stage of preparation by RSSB, and
further general information is available from the British
Psychological Society website (see Further information at
the end of this section).
Biodata (biographical data) questionnaires
This method takes the form of a multiple-choice
biographical questionnaire and is more likely to be used
by large organisations. Sometimes applicants may seem
much the same on paper, but some have greater initiative
or ‘people skills’ than others. Biodata consists of systematic
information about hobbies, interests and life history, which
is assumed to be more or less indicative of the suitability
of people for different kinds of jobs. The main use of
Understanding Human Factors/June 08
Part 2: Guidance
Staffing
biodata is in the pre-selection of people for basic-level
jobs, such as apprentices or graduate trainees. The logic is
that if candidates are matched with existing staff, people
with similar interests can be found who are likely to be
suitable for the job. The greatest value of the technique is
its ability to reduce staff turnover.
Assessment centres
This is often a day-long, or even residential, process in
which candidates are put through a battery of interviews,
psychometric tests, group discussions and other exercises.
A well-known example is the in-basket exercise. Here,
applicants work through a pile of written notes and
problems and must determine how – and in what order
– they would handle each. In-basket exercises are useful
to determine how applicants work under time pressure
and how they plan their time, as well as their ability to
distinguish priorities.
Which selection method is best?
There has been a lot of research into the effectiveness of
the different methods of selection. By ‘effectiveness’ we
mean how well the results of a selection method predict
later performance on the job. This is sometimes known as
‘predictive validity’.
The ability of a method to predict later performance
depends on the extent to which what it measures turns
out to be necessary to the performance of the job. For
example, a selection method might include a test of
intelligence or verbal ability, both of which are clearly
important to all jobs in the railway industry. However,
training or job experience may be equally important. If
they are not taken into account the utility of the selection
Understanding Human Factors/June 08
method will suffer. It will
also suffer if the method
turns out to be a bad test
of what it is supposed
to be measuring. For
example, there are good
and not so good tests of
intelligence.
Which selection method is best? (Listed in order of effectiveness)
Method
Best for
Cost
Effectiveness
Assessment Centre
Applicants to higher paying and/or higher
risk positions, where the consequences of
a wrong hire decision may carry higher
penalties for an organisation or its customers
High
Good
Psychometric tests
of intellectual ability
Applicants with no previous experience in
the job for which they are applying. Here,
the selection method is used essentially to
measure the ‘trainability’ of the applicants
Moderate
Moderate to Good
Practical tests or
Applicants with previous experience of the
Moderate to High Moderate to Good
The Panel, Which
trials
job they are applying for
selection method is best?,
Structured
Applicants with previous experience of the
Moderate
Moderate to Good
summarises the general
competency
job they are applying for – although readsituation. Note how
interviews
across from similar situations is often possible
far down in the list the
Previous job and/or All applicants
Low
Moderate to Good
personal references
traditional structured
Biodata (biographical Applicants with little previous experience and Low
Low to Moderate
selection interview
data) questionnaires who will be new trainees
comes. This is because
Structured
All applicants where it is not appropriate or
Low
Low
the selection interview
traditional selection
practical to carry out structured behavioural
has high face validity, but
interviews
interviews or practical tests
low predictive validity. In
Psychometric tests
Do not use on their own – may have a
Low to Moderate Low
of personality
helpful role as part of an assessment centre
other words it may seem
Do not use!
Low to Moderate Nil
Unvalidated tests,
to be powerful, but it is
eg graphology
not a very good measure
(handwriting),
of how well a person will
astrology etc
actually do on the job.
performance. The best psychometric test or practical
The selection interview is
test only measures a small percentage of the reasons
important – but should not be used on its own to make
why people do well or poorly at their jobs. When the
reliable selection decisions.
best methods are combined – as they are in assessment
centres – the percentage gets bigger. However, even then
In choosing a method (or a combination of methods),
it is still the case that most of the reasons for successful
you will need to take into account how appropriate they
job performance cannot be assessed. This is because job
are to your situation and resources. But remember,
performance is affected by a large number of interacting
no selection method on its own – or in combination
reasons that emerge as job holders develop over time: eg
– provides anything like a certain prediction of job
their changing domestic circumstances; their relationships
Page 81
Part 2: Guidance
Staffing
at work; their hopes,
ambitions and fears; and,
most of all, their experience
of doing the job itself in the
context provided by the
culture of the organisation. As
time goes on, the relationship
between selection test results
and job performance gets
weaker and weaker.
‘The traditional interview
is not one of the best
selection methods…but
you can improve its
usefulness by following a
basic few rules.’
The real value of selection methods is that when they are
applied appropriately, with care and in combination, they
can sufficiently predict performance for the next two to
three years to be capable of saving organisations a lot of
time and money.
How should you conduct a selection interview?
If you are recruiting to a safety-critical role, the traditional
selection interview will form only part of a selection
process, in which HR professionals should also be
involved. The interview is, however, the most accessible
method around, and for this reason alone it is here to
stay. You can greatly improve its usefulness by following a
few basic rules.
In essence, you need to:
• Plan the interview properly – to make things as easy
as possible for you and as fair as possible for the
applicant. This will involve properly examining whatever
materials the applicant has sent you, and being very
clear about what you are looking for
Page 82
•Conduct the interview fairly and
consistently for each applicant – to
give yourself a basis for comparison.
Remember that there is very often a poor
correlation between your first impressions
and the end of the interview – let alone
performance on the job (see Panel, Don’t
discriminate!).
•Follow-up the interview promptly – to
give feedback to the applicants and to
ensure that references are followed up before your
preferred applicants are snapped up by someone else!
Before interviewing
1 Spend time with the results of the task analysis (page
47). Use it to develop a list of questions that will
answer ‘What relevant knowledge, skills and attitudes
do the applicants have?’
2 Sift applications. Do not arrange to see too many
people for one job – you will be wasting everybody’s
time, including your own. And do not plan too many
interviews on one day. Four or so is about right.
Don’t discriminate!
In using any method of selection, you will need to
take great care not to fall foul of the discrimination or
data protection laws. The discrimination laws make it
unlawful to discriminate on the grounds of someone’s
sex, sexual orientation, marital status, race, colour,
nationality, ethnic origin, religion, beliefs or because of a
disability, pregnancy or childbirth, or membership/nonmembership of a trade union. It is also unlawful to
discriminate against part-time workers. A good rule of
thumb is to avoid probing personal or private topics
that have no relationship to the candidate’s ability to
perform the job. With regard to the Data Protection
Act (1998) you should remember that candidates have
the right to demand all the data that you collect about
them – so be careful what you record!
6 Book a quiet place for the interview where you will
not be interrupted. Try to arrange for a round table
rather than a desk (often perceived as a barrier).
7 If you plan to have other individuals in the meeting,
plan your roles in advance, ie who will ask what
questions.
3 Review each applicant’s resumé and application form.
Write down questions that will answer ‘What more
do I need to know about this particular applicant?’
4 Make sure you have a written job description and
that you are familiar with it so you can answer the
applicant’s questions.
5 Be aware of the kind of questions you must avoid
asking (see Panel, Don’t discriminate!)
Understanding Human Factors/June 08
Part 2: Guidance
Staffing
The interview
1 Make sure you are there before
the applicant and that you are
relaxed and prepared – remember
that the applicant is judging you
and your organisation as well.
Make sure reception staff are
expecting the applicants.
‘Remember – if you
think the applicant is
good - someone else
will too … keep in
touch with them while
you are making sure!’
2 Greeting the applicant in a friendly,
welcoming way. Break the ice by (eg) asking where
they have travelled from and how their journey was.
Do not imagine that you are able to sum up the
applicant in the first few minutes of meeting them –
you will almost certainly be wrong.
3 Do not get straight into questions. If there is more
than one interviewer present, make sure you explain
to the applicant who the other people are and why
they are there. Explain the purpose of the interview
and outline the organisation and how the job fits with
the organisation. Ask if the applicant has any questions
about the job or the organisation before beginning
settling into interview.
4 Start with straightforward factual matters covering
things like applicant’s full name, address and current
employment status, employment history etc. They are
useful to check and easy for the nervous interviewee
to answer.
5 As the interview proceeds, make sure you give the
applicant time to think about your questions and
to consider an answer. Do not interrupt – but do
intervene if the applicant becomes flustered. Use
Understanding Human Factors/June 08
silence after you’ve asked a question.
You’ll get interesting information if
you let the applicant fill the gaps you
leave. The applicant should be talking
75% of the time. Any less and you are
talking too much! Keep control of the
interview. If you feel the candidate is
going off-track, turn the conversation
back to the information you need.
6 While interviewing the applicant, keep eye contact
and don’t anticipate answers, but do listen attentively.
And don’t show whether you agree with or approve
of what an applicant tells you.
7 Use open-ended questions to get information. They
begin with What, Why, How, Describe or Tell me about.
Use behavioural questions where possible (see Panel
on page 84, More about competency interviews). Use
closed questions (that can be answered Yes or No) to
confirm information.
8 Ask only job-related questions and make notes during
the interview – but be aware of the Data Protection
Act implications (see Panel on page 86, Don’t
discriminate!). Make sure you develop a picture of how
the interviewee’s knowledge, skills and attitudes fit
with the job description produced by the task analysis
(page 47). Try to get an understanding of what they
value about their work and their relationships. A good
question is to ask ‘what were you most proud of in your
last job and why?’ or ‘what did you find most challenging
and why?’. Make sure you understand how they
come to be applying for this job at this time, what
explanations there are for any historical gaps, and
where they see themselves going in the future.
Following up on references
References are so important because a person’s track record is probably
the most valuable indicator of future success. You need to be sure
that the people you hire can do the job and have nothing to hide that
might endanger the workplace – or your organisation. You should ask
the applicant to give you at least two work-related references whom
you can call to discuss their experience of working with the applicant.
Ideally these should be people to whom they reported in previous jobs.
If the applicant can’t give the name of a previous boss who will act as a
reference for them, you need to understand why.
Tips for conducting reference checks
•If possible, get candidates to ask their referees to call you. The referee
will be much more willing to share information with you if they were
asked by the candidate to do so and the call comes from them.
•Where this is not possible (or you get no response within a few days),
send a simple written form for them to fill in. Do not make it arduous
or you will decrease your chances of a reply. Give them the option of
calling you if they would prefer.
•Keep your questions consistent from reference to reference so you
can compare responses.
•Pay more attention to negative or neutral information compared to
positive information. Most people find it easier to say good things
about an ex-employee than bad things.
•Keep your questions focused on what you need to know about
the candidate’s performance in previous jobs. Good questions for
references include:
•How long have you known the candidate?
•What was your reporting relationship to them?
•How would you describe their contributions to your organisation?
•How would you describe the candidate’s strengths? What skills could they
further develop?
•How did they get along with others?
•What seems to motivate them?
•Would you re-hire them?
•Is there anything else you’d like to add?
Adapted from Reed (2003), reproduced with permission
Page 83
Part 2: Guidance
Staffing
After the interview
Recruitment
1 Ask the applicant what remaining questions they have
- but don’t conclude anything from a nil response.
How can you recruit the right people?
2 Sincerely thank the applicant for coming and let them
know when they are likely to hear the outcome. Do
not hint what the outcome will be.
3 Make a written record as soon as possible after
the interview. Only record what was said in the
interview. Do not record your beliefs or thoughts
about the applicant – records may have to be made
publicly available later. Only keep personal data after
an interview if it is necessary and relevant to the
selection process, or in respect of a discrimination
challenge. All such data must be securely stored.
4 For promising applicants, take up references from
former employees. References are a very important
source of information about a prospective employee,
(see Panel on page 83, Following up on references).
Let each applicant know if they are going to be
considered. Tell them how long it will take for a
decision to be made - and be as accurate as possible.
5 Make a decision by comparing all of the evidence
you have gathered for the applicants, including the
references information, and setting it against the job
description, together with your knowledge of the job
requirements and present staff.
6 Inform applicants of the outcome by the date you
promised.
Page 84
There are two levels of answer to this question. The
first level is at the ‘sharp end’ and concerns the channels
by which your organisation can find potential recruits.
The second level is a more strategic one and concerns
the whole issue of what can be done to make potential
recruits want to work for your organisation.
How can you find new recruits?
Most organisations are familiar with the main recruitment
channels, which include the following:
• Recruiting from within. Your first action should be to
see if you can fill a vacant position from your own
staff. Perhaps someone needs promoting, and both
the employee and the organisation already know each
other well. Recruiting from within is also good for
employee morale (page 117), as it shows the rest of
the workforce that progression is possible. It is also
usually easier to fill the lower level position vacated by
the move. (See supervision and appraisal on page 68
for issues and strategies to cope with this.)
• Staff networks. If the post cannot
be filled directly by existing staff, it
is quite likely that potential recruits
are known to those same staff, ie
friends and relations. These staff
networks should also include all
those former employees whom you
were reluctant to lose, but whose
circumstances may now have made
them receptive to re-employment.
• Bounty rewards. It might be possible for your
organisation to operate a bounty reward scheme for
staff who recruit people to the organisation. Bounties
can vary depending on how difficult it is to find people
to fill the position. Even though the bounty involves
spending money, the amount will be a lot less than an
agency fee and represents high value: it not only fills
a vacancy, but spreads good will since someone from
your own staff will benefit.
• Temporary personnel agencies. Temporary employment
agencies are a flexible source for a wide range of
employee types. It might seem an efficient way of
finding suitable new recruits – but there are pros and
cons. The advantage is that you certainly can evaluate
people for the possibility of permanent hire without
committing yourself. The disadvantage is that the
agency will probably charge if you do hire them.
• Employment agencies. These agencies will find you
permanent employees – for a fee. They will usually
have a suitable portfolio of potential recruits – and
should have already checked their references, which is
a crucial step in your hiring decision.
• Executive search firms. Also known as headhunters,
are more appropriate for senior
‘Recruitment is a twoappointments. You can expect to pay
way track … you need a substantial fee for a successful hire.
to do everything you
can to make it you
that’s wanted by your
potential applicants!’
• Your organisation’s website.
Your website is an obvious – even
expected – recruitment channel.
You should ensure all your open
positions are posted on it. This
Understanding Human Factors/June 08
Part 2: Guidance
Staffing
Make sure people want to work for you!
Whatever the medium for your recruitment
communication, you should never assume that your
organisation is in the driving seat because it’s you that
has the job(s). There is always keen competition for
the better recruits. Recruitment will be a lot easier if
it you can project an image that makes people come
to you. Your organisation can do this strategically. But
you, too, can help to create an attractive target for your
potential recruits by following some basic ground rules
when you publish recruiting notices. In particular, you
should consider the following:
• Include something unique about your organisation
• Present an appealing and accurate representation of
your organisation
• Use current employee endorsements
• Declare what education and training will be given, if
applicable
• Be clear any required education and training
• Include required previous job experience, if
applicable
• Don’t over-specify – if a skill is not essential, don’t
ask for it
• Don’t run foul of the law (see Panel on page 82,
Don’t discriminate!)
channel has the added advantage that people
who use it is likely to be both well-informed and
realistically-motivated. In effect, they will have prescreened themselves, as long as you have published an
Understanding Human Factors/June 08
appropriate amount of detail about your organisation
and the jobs available.
• Want ads. Want ads are a traditional way of getting
your message to a great number of potential recruits.
The obvious channels for want ads are the various
trade, industry and railway publications printed both
weekly and monthly.
• Employment websites. The Internet now contains
a great number of third-party sites that carry job
postings, want ads and resumés of prospective recruits.
A good way to find these sites is through Internet
search engines, such as Google™.
How can you encourage recruits to find you?
One of the rail freight companies currently franchised to
operate in the UK hardly ever needs to recruit drivers.
The drivers find the company. They hear about the
modern, innovative working practices the company offers
through word of mouth. They check things out by visiting
the company’s website and they call.
include implementing modern working practices,
schedules and management structures (like the rail
freight company), participation in best practice staff
development and training programmes such as the
Government’s Investors in People programme, or
operating a caféteria-style benefits programme (so
that employees can select the package of benefits that
best suits them).
• Get involved in community events and programmes. Local
communities – and sometimes, not so local ones –
respond well to organisations that spend resources on
helping them to address some of the problems they
face. This can represent a good investment – especially
if it is combined with an open-door recruitment policy
in which people are welcomed to find out more about
employment prospects.
• Sponsor work-study programmes and create good
relationships with schools careers officers. These sorts
of initiative spread the word very effectively, leading to
natural employment opportunities for school leavers
– leading in turn to positive attitudes amongst the
parents of those school leavers, and so on.
There are several ways in which organisations can do
a lot strategically for their recruitment prospects. The
important thing is to promote a
‘An organisation’s ability to
positive image for the organisation
that has high visibility – both inside
recruit the right people in
and outside the organisation – with
the right numbers at the right
initiatives such as the following (see
Panel, Make sure people want to work
time is intimately connected
for you!):
• Provide a working environment
that employees value highly –
the word will spread. This may
to its strategic and economic
decision making.’
• Open up your gates for
classroom visits and possibly
Public Open Days for a
‘behind the scenes’ look. This
will be more appropriate
for some companies than
others (eg those with
engine yards, control rooms
and signal boxes etc). While
there are always health,
Page 85
Part 2: Guidance
Staffing
safety and security matters to consider, it is the case
that the railways hold a fascination for many people.
Using this fascination to build rapport with a large
community can produce significant visibility.
How many people is enough?
This is not just an economic question, but one of safety
as well. An organisation needs to be sure that it can
meet the demands of its customers in a way that keeps
it profitable. But it also needs to be confident that it can
meet the demands of severely degraded or emergency
conditions and so fulfil its safety responsibilities.
The question of how many is enough is easier to answer
in some areas of the railway industry than others.
Where there is a clear staffing requirement (eg a train
timetable, a signal box with a specific number of panels
or a depot with a specific number of service bays) and
an agreed shift work pattern (page 128), it is relatively
straightforward to work out how many people are
needed to make the system work. On the face of it,
the impact on recruiting of complicating factors, such as
sickness rates, holiday entitlements, train failure rates and
servicing schedules are also relatively easy to calculate.
Difficulty for recruiters can, however, start to occur
when some of these ‘fixed’ assumptions turn out to be
not so fixed. For example, it is tempting for a TOC’s
accountants to re-think the maintenance regime for
its train fleet, and cut out one of the major services by
increasing the amount of maintenance work done during
minor services. While this will save significant amounts of
rolling stock downtime during the major service, it means
that the number, level – and therefore cost – of fitter
Page 86
How can you tell if your organisation is the wrong size?
There are several things you can do to establish whether your organisation is the wrong size:
• Benchmark staffing levels. On paper divide all your staff into two groups – those who actually earn the revenue for
your organisation and those who are employed to support the revenue earners. For both groups, add up all the
weekly hours, including their typical overtime, and divide by 40 to calculate the number of full-time staff members
in each group. Divide the support group number by the revenue group number to calculate the number of
support staff you employ per revenue earner. Now compare that number with other organisations like yours,
including those considered to be ‘best performers. You may find that the best-performing organisations have
higher support staff-to-revenue ratios and typically spend more on support staff. But support staff costs may be a
smaller percentage of total revenue, because the best performing organisations bring in more income.
• Track productivity. Keep tabs on a few staff for a few days to see how many tasks they perform, then compare
those figures to the norms elsewhere. But don’t use this information as the basis of hiring/firing decisions – it
would be very premature. Instead, use the data to build organisational snapshots that help you spot problems. If
your organisation’s productivity seems low, try to find out why. Perhaps existing or new staff members need more
or different training. Or maybe your computer system is not working properly. Or perhaps a working practice has
not kept up with recent changes. You might find that staff members are being pulled away to do other tasks, as in
the case of a station despatcher who is needed elsewhere on the station.
• Look for signs of ‘wrongsizing’. If you have too few staff members, you are probably paying a lot of overtime, which
affects profits and burns out staff. Another classic sign of trouble is that staff can’t stay ahead of train movements
or maintenance schedules. Even organisations that have enough staff can often do a better job with staff
scheduling. Do you never have enough station staff or fitters during peak times, but find staff standing around at
other times of the day? In that case, adjust staff schedules and shift patterns to better match demand, or hire parttimers to get the coverage you need.
• Simplify your workflow. Staffing needs are influenced by how complex you make your organisational rules and
procedures. While care must be taken to ensure that safety is not compromised, it is sometimes the case that
rules are put in place more to cover the organisation than to facilitate its work – see Why do people break
rules? (page 12) Create a flow chart for each staff position or department, which shows the steps needed to
accomplish daily tasks. Identify steps that create barriers or do not add value, and look for simplifications.
• Audit staff activities. If your most experienced staff spend their time doing things below their skill levels, you’re not
taking advantage of their expertise – and they are getting de-motivated.
Adapted from McGuire (2002), reproduced with permission from ACP Online.
Understanding Human Factors/June 08
Part 2: Guidance
Staffing
skills must be greater for minor services. Simultaneously,
work will be taken away from the TOC’s maintenance
contractor, who has arranged their business so that they
can perform a certain frequency of major services. If
this demand is removed, the contractor capability will
probably vanish with it. In a case such as this, changing
an assumption will mean that the recruiters in both the
TOC and the contractor will have provided the wrong
numbers of people with the wrong type of skills at the
wrong times. And this will have major consequences for
both organisations – in terms of staff morale (page 117),
economics and future possibilities.
Retention
How do you keep the right people?
Why is retention important?
For most situations, the keys to employee satisfaction and
retention hinge on:
Losing people is expensive. That said, retention is actually
something of a bidirectional line. It is an essentially
• how valued they feel
desirable goal – since the recruitment (page 84) and
• how challenged they are
selection (page 79) processes are expensive and divert
organisational resources from earning revenue. In addition,
• what opportunities they have for growth and
the hiatus that can be caused while the organisation and
advancement.
new staff adapt to each other can be very disruptive
to overall productivity (see
Many of the strategies aimed
supervision and appraisal on page ‘Listening is the most costat attracting prospective
68). However, staff turnover
effective way to acknowledge employees to an organisation
is also an essential part of the
It is important to understand how numbers and quality
in the first place will, of course,
people. Being heard builds
process by which organisations
of staff affect, and are affected by, decisions elsewhere
be effective in helping to retain
renew themselves with different
in the organisation. A general technique for doing this is
self-esteem, and employees
them. This means that many
thinking and ‘new blood’.
the influence diagram (see cognitive mapping in Part 3,
of the strategies that are good
with high self-esteem feel
Techniques). Based on the premise that things are in the
for recruitment (page 84) will
The ideal situation is one in
end connected to everything else, the influence diagram
trusted
and
valued.
’
also support retention. These
which staff only ever leave
is a simple but powerful means to track the knock-on
Mike Flaherty
strategies will be complemented
effects of decisions as they ripple through an organisation. an organisation for positive
by an effective selection process
reasons – and the organisation’s
that can pick the right people for the organisation’s jobs
strategic recruitment plans allow for this. Such plans
Research has also revealed indicators that allow an
and culture. Much of the best practice that you’ll find in
may even deliberately allow for personnel swaps with
organisation to detect when its recruiters are getting it
the section on morale and motivation (page 117) applies
other organisations or industries, and for sabbaticals,
wrong – or perhaps not being allowed to get it right (eg
here.
in recognition that key staff will need to be developed
by their finance department or company strategists). As
in ways that single organisations cannot provide.
it turns out, ‘rightsizing’ an organisation may lead to more
The following best practice is particularly relevant to
Arrangements like these are mutually beneficial and
staff (and therefore more operating costs)
retention
help to assure the
but ultimately more profit and more safety
‘The average cost of replacing recruitment of
• Listen to your staff. Flaherty (2002) says that “listening
– through less staff turnover, less disruption
valuable
people
in
is the most cost-effective way to acknowledge people.
an employee is between 1
and better capacity to deal with degraded
the
first
place.
Being heard builds self-esteem, and employees with
or abnormal working (see Panel on page
and
2.5
times
the
employee’s
high self-esteem feel trusted and valued”. The main
86, How can you tell if your organisation is the
means by which you should listen is through your
wrong size?)
annual salary plus benefits.’
supervision and appraisal process (page 68). Best
The Gartner Group
Understanding Human Factors/June 08
Page 87
Part 2: Guidance
Staffing
practice here will result in an honest assessment of the
‘fit’ and potential of staff – as well as the identification
of a development path that makes the most sense for
both the individuals and the organisation. If individuals
see that they have a long-term future, they are likely
to stay around. And if they choose to go anyway, then
it’s a good idea to listen to them as they go out of the
door – so, make sure you conduct an exit interview.
• Train your staff. Conducting regular training needs
analysis (page 55) – usually as part of the supervision
and appraisal process (page 68) and leading to
organisational investment in staff training (page 55)
– sends a powerful message to everyone about how
much you value your staff. Investment in training is also
a vital tool at other times. Organisations sometimes go
through difficult times, which can lead to redundancies
and re-organisation. It is likely that the employees
you most care about and who escape a layoff won’t
sit around very long wondering if their jobs are safe.
Scheduling training for them can provide much-needed
reassurance at times like this.
You should also consider how you can use the
knowledge that your staff has already acquired for the
benefit of others. People who have worked in your
organisation for a number of years are experts. Make
sure they are involved in the induction programme for
new recruits. There may be other opportunities, too eg on public open days (see Recruitment on page 84).
Their knowledge will be useful and you will make them
feel valued.
Page 88
• Offer incentives. Employees are more committed when
there is a financial reward at stake. Paying attention
to this aspect of retention is an additional reason for
a recruitment bounty reward (see Recruitment on
page 84). If ‘employee of the month’ seems a little
too brash for your organisational culture, it shouldn’t
take you too long to think of more appropriate staff
loyalty schemes. For instance, you could arrange an
incentive that adds up to an additional 10% of their
base pay as a bonus. Of that 10%, 60% might be linked
to individual performance; 30% to their team goals; and
10% to the entire organisation’s performance.
More informal, non-financial reward systems can
also be very effective. Here, recognition is linked to
personal desires such as time off, job sharing, flexitime,
office space, special tasks, public acknowledgment,
news releases, etc. The most important part of any
informal reward/recognition system is that it is linked
to organisational values and that it is seen to come
directly from management.
• Don’t reward the wrong people. Don’t reward poor
performance by giving overtime to someone who
works too slowly. And don’t penalise good performers.
It may seem easier to make allowances for less
efficient staff – while failing to extend the same
flexibility to people whom you depend on to get
things done. It’s also tempting to keep piling projects
onto high achievers, making them work much more
than less productive colleagues. These are mistakes that
will lead to you losing your best people and retaining
your worst.
Further information on staffing
1 BPS psychological testing website, British Psychological
Society www.psychtesting.org.uk/ (as of May 2008)
2 Cronbach L.J. & Gleser G.C. (1965) Psychological tests
and personnel decisions, Urbana, Uni of Illinois Press
3 Dukes C. (2001) Easy Step by Step Guide to
Recruiting the Right Staff, Rowmark
4 Flaherty, M (2002) Rural Institute article on staff
retention, University of Montana, see - http://
ruralinstitute.umt.edu/training/publications/
newsletters/staff_retention.asp (as of May 2008)
5 Herriot P. (1989) Recruitment in the 90’s, Institute of
Personnel Management
6 McGuire, P. (2002) Rightsizing, not downsizing is key
to staffing success, American College of Physicians,
www.acponline.org/clinical_information/journals_
publications/acp_internist/may02/rightsize.htm
7 Pell A. (2000) The Complete Idiot’s Guide to
Recruiting the Right Staff, Alpha Books
8 Reed, C. (2003) Recruiting the Right People, part of
the Human Resources Management
Tools Project by NetGain Partners
Inc. for the Cultural Careers Council
and the Cultural Human Resources
Council, Ontario, Canada
9 RSSB (2008) RIS-3751-TOM Rail Industry Standard
for Train Driver Selection Issue 1 Aug 2008, RSSB,
Euston, London
Understanding Human Factors/June 08
understanding
© Four by Three 2003
Culture
human factors
understanding
Culture
human factors
© Four by Three 2003
Part 2: Guidance
Culture
Culture
An organisation’s culture is a complex, subtle
phenomenon that influences every aspect of
organisational life. This means that staff at every level have
to be sensitive to their organisation’s culture if they are to
achieve today’s goals and bring about change in readiness
for tomorrow’s challenges.
Focus on culture
Understanding Human Factors/June 08
The diagram on this page focuses on culture. It shows
five sub-areas (in the middle red ring), and identifies the
main human factors questions that this Guide answers (in
the outer grey ring). At the end of each section, you will
find a list of sources of further information. But first, we
provide a concrete introduction to what can be a rather
vague notion.
What is organisational culture?
People in the railway industry – as in all
organisations –
sometimes make mistakes and break
rules (see Why do people make
mistakes? on page 9) and Why
do people break rules? on page
12), but most of the time
they do what is expected
of them. This is partly, of
course, because they want
to get paid and because, in
safety critical areas of work,
the Rule Book defines in
considerable detail what
must be done. However,
for much of the time,
people do not work with
either their pay or the Rule
Book consciously in mind.
Instead, they work in response
to a set of shared goals, values,
beliefs, expectations and so on,
that they have acquired during their
career in the railway industry. It is these
that, together, comprise the organisational
culture of the railway industry as a whole.
Of course, the situation is complicated by the fact that
different companies and different occupational groups
also have their own unique culture.
Usually people do not talk about the values, beliefs, and
so on that they share with other people working on the
railways. But the fact that they are shared means that staff
can usually work together without major difficulties, even
when they don’t really know each other. It is only when
somebody starts acting in accordance with, say, different
values or beliefs, that you may become aware of what
you normally take for granted.
How do you recognise what the culture is?
In order to ‘read’ your organisation’s culture, you need to
be able to decipher what messages are being sent at each
of a number of levels – namely:
• how things and people appear
• what people say they think and believe
• what people really think and believe.
Based on: Schein (1999)
How things and people appear. These include not only all
the physical things – such as architecture, room layout
and decoration, computers, car parks, reception areas
and messrooms – but also things to do with the people,
such as dress code, uniforms and general tidiness of
appearance. – as well as the way they behave. They also
include certain jobs that say something about the kind of
organisation it is – such as receptionists, security guards
and car park attendants – and working practices. These
cover what people do and how they do what they do.
Are they, for example, brisk and business-like? Do they
Page 91
Part 2: Guidance
Culture
appear bored and slow or even
reluctant about their jobs? Are
they quiet or chatty? Do they talk
about the weekend’s football or
only about work?
‘Culture is the meaning
people share for the things
they can see, the values they
have and the goals they
pursue. Culture is a two-way
track: it emerges from the
things people share ... and it
allows people to make sense
of what is shared.’
What people say they think and
believe. There is often a mismatch
between an organisation’s stated
values and the values reflected
in what it does. For example, an
organisation may state that it
values good working conditions
for its staff. But in reality it may
have dingy messrooms, dirty
washrooms, old and poorly maintained company vehicles,
and so on.
What people really think and believe. These are basic
assumptions which are rarely, if ever, spoken, about the
way the world really is. They are held about questions
concerning the nature of reality, business and work,
people and much else. For example, is the world of
work a harsh, cut-throat competitive environment in
which only the strong will survive? Or is work best seen
as a cooperative enterprise in which everyone must
work together? Are people basically lazy and can they
only be made to work in a regime of close monitoring
coupled with tightly defined systems of incentives and
punishments? Or do most people actually want to do a
good job, and just need the right resources, direction and
encouragement?
Page 92
The assumptions widely held
within an organisation about
the answers to these and many
other questions will ultimately
determine how decisions are
made, who gets recruited and
promoted, how resources are
allocated, how staff are treated,
and how customers, suppliers
and competitors are viewed
and dealt with. In short, they
will determine how the whole
organisation functions.
Teamworking
A central part of any organisation’s culture is how people
relate to, and work with, each other. How an organisation
uses teams and how these teams work are key cultural
issues.
Because organisational culture
is all-pervasive, we could consider any of the topics
covered in this Guide in relation to it. Here, however, we
concentrate on five areas where culture is more of a
foreground issue.
Change
Any organisational change is only true change if it lasts
and transforms the nature of the organisation in some
way – or, in other words, involves a change in culture. This
is immensely difficult and takes a very long time.
Leadership
Leaders have an important role in forming the
organisational culture. While most organisations in the
railway industry have formally designated leaders, in
practice every member of staff is probably called upon to
exercise leadership at some time or another. This means
that everyone, regardless of their formal position, needs
to have some understanding of how to be a leader.
We also pay particular attention to safety culture. It is
important to know how this manifests itself in your
organisation, how it can be managed and, where
necessary, changed.
Management
Being a manager is not the same thing as being a leader.
But managers are obviously responsible for ensuring that
the necessary things get done in the organisation, and
that they get done properly. In this section we will look at
some key cultural aspects of management.
Communication
People in almost every organisation report
communication as being a major problem. What is
communicated, by whom, to whom, how and when, are
important aspects of the culture. But communication is
not only a manifestation of the culture; it is also the main
mechanism for spreading and institutionalising the culture.
Why do you need to be concerned with your
organisation’s culture?
A good culture plays an important role in organisational
success. When there is a good culture, people have little
uncertainty: they know what is expected of them, what
they should do, and what they can expect of others.
Avoiding uncertainty means avoiding confusion and delay,
two of the key ingredients in poor performance and poor
safety.
Understanding Human Factors/June 08
Part 2: Guidance
Culture
Having awareness and knowledge of your organisation’s
culture means that you can work with it rather than
against it. The values and beliefs that form the bedrock of
culture have evolved because they work – at least most
of the time. They can be an important guide to action
when there are no rules or procedures to direct you, or
in novel situations when the existing rules and procedures
no longer apply. Those around you are more likely to
act on any decisions you make if these decisions are
consistent with their expectations.
Having knowledge of your organisation’s culture – both
its strengths and limits – will also play an essential role in
helping you to bring about change. It’s important to bear
in mind that true organisational change almost always
entails some change in culture.
What is safety culture?
In this Guide we pay particular attention to safety culture.
Safety culture is the set of values and priorities placed
on all aspects of safety by everyone at every level of an
organisation. Although the concept of safety culture was
widely used in other industrial sectors and was already
known within the rail industry, this idea has been given
much greater prominence since Lord Cullen addressed
safety culture on the railways so forcefully in his inquiry
into the Ladbroke Grove accident. Since then, RSSB has
produced a number of reports addressing various aspects
of safety culture and the closely related concept of safety
climate (see Panel, Definitions).
The interest in safety culture comes from the
realisation that no matter how good the engineering or
technological measures to ensure safety on the railways,
Understanding Human Factors/June 08
these will not be effective if safety is not a key value that
exists at all levels of railway companies.
Each of the five areas of culture that we consider in the
following pages – leadership, management, teamworking,
Definitions
Safety culture
A combination of values, beliefs, vision, purpose,
policies, objectives and leadership styles that impact
on an organisation’s safety. A positive safety culture is
characterised by awareness, assessment and action on
safety matters in all these areas, and is supported by
an open communications style throughout the whole
organisation.
Safety climate
A snapshot of the surface features of the safety
culture resulting from the workforce’s attitudes and
perceptions at a given point in time. Sustainable change
is achieved through continually looking beneath the
surface and questioning assumptions.
It is helpful to think of culture and climate as the invisible
and visible portions of an iceberg. Safety climate, the
visible part of safety culture, only represents the tip
of the iceberg. The actual problems of culture might
lie hidden deep beneath the surface. Rather than just
treating the visible symptoms, real improvements can
only be made by changing the underlying culture.
Best practice is to develop excellent communications
and teamworking, and early warning systems, eg near
miss reporting within your organisation.
Adapted from: Rail Safety, Safety Culture Bulletin 1, Jan 2002
‘Safety culture is the set of
values and priorities placed
on all aspects of safety by
everyone at every level of
an organisation.’
communication and change – has something important to
say about safety culture.
A wide range of tools and methods are available for
assessing safety culture. RSSB has developed a webbased Safety Culture Toolkit that will allow rail companies
to measure their own safety culture, and determine
what actions they could take without the need for
extensive external support. The Toolkit facilitates the data
management, making it easier to establish a single industry
view. It also enables the benchmarking of individual
companies’ cultures. See Further information.
The Panel on page 94, How should you assess safety
culture?, explains how you should approach the
assessment of safety culture in your organisation. (Note
that all safety assessment processes should be designed in
accordance with HS(G)65 – see Further information.)
But what should you be assessing?
HMRI have recently published the safety culture
inspection toolkit. Although designed for use by HMRI
inspectors, the approach taken is of much wider value.
The toolkit is based around the assessment of five safety
culture indicators (see Panel, What should you assess?).
Page 93
Part 2: Guidance
Culture
How should you assess safety culture?
What should you assess?
Plan the assessment
Planning is everything, and communicating the plan and providing
feedback are equally important.
Leadership
Management must take explicit and continuous steps to ensure that
goals, targets and issues are made clear, and are known to all personnel.
An indicator of good safety leadership is that safety is always prioritised
over performance.
Recognise the importance of involvement
Involve staff throughout the whole assessment process. This will help
them to buy-in to the outcomes from the assessment.
Ensure effective communications
Communicate management’s commitment to the process so that
the staff have confidence in the importance of the process and that
something will be done with the results. Provide plenty of information on
the reasons for doing the assessment and how the results will be used.
Provide feedback
Do this as soon as possible after completion of the assessment so
that staff see the momentum is being maintained. If you are providing
feedback following introduction of improvement actions or other
changes, make clear what the changes have been at the beginning of the
feedback.
Analyse results
Use a form of analysis that will produce meaningful results to enable
tracking of future changes and further feedback to staff.
Review and discuss the issues raised and request clarifications from each
group that has taken part in the assessment.
Create an effective action plan
The action plan should start by addressing the most significant or critical
development needs. Plan re-assessments as part of the action plan, but
do not be tempted to reassess for between 18 and 24 months to allow
changes to take effect.
Source: RSSB (2005)
The content of this panel is highly consistent with the
requirements for High Reliability Organisations (HROs).
Research has discovered that HROs have a safety culture
that is different from other organisations in a number
of important ways. The result is that they have far fewer
incidents than might be expected, given the dangerous
Page 94
Two-way communication
There are multiple channels for the discussion of safety matters, concerns
and goals between and within all levels of the organisation. The flow of
information should be in an upwards as well as a downwards direction.
Employee involvement
Personnel from all levels within the organisation should be involved in
decision making, safety planning, and providing ideas for improvement.
Employee participation and feedback should be actively sought.
Learning culture
Steps should be taken to monitor known problems, identify new ones,
detect trends over time and develop effective preventative measures.
Efforts must be made to ensure that lessons are learned from incidents,
including the wider application to other situations. Intervention measures
must be introduced for all situations.
Attitude towards blame
Developing a just culture is the acceptance that the ultimate
responsibility for incidents lies with the organisation, and investigations
must therefore take full account of multi-causality. The purpose of
investigations is not to take retribution or assign blame, but to learn from
incidents.
Adapted from: HSE (2005), reproduced with permission. © Crown copyright
material is reproduced with the permission of the Controller of HMSO and
Queen’s Printer for Scotland.
nature of their operations (see Panel on page 95,
How does a high reliability organisation differ from other
organisations?)
Further information
1 Deal T & Kennedy A. (1988) Corporate Cultures: The
Rites and Rituals of Corporate Life, Penguin
2 Helmreich R.L. & Merritt A.C. (1998) Culture at Work
in Aviation and Medicine: National, Organizational and
Professional Influences, Ashgate
3 Hofstede G. (1994) Cultures and Organisations
– Intercultural Cooperation and its Importance for
Survival, HarperCollinsBusiness
4 Schein E.H. (1999) Corporate Culture – Survival
Guide, Jossey Bass
5 HSE (no date) Successful Health and Safety
Management, HS(G)65, HSE Books
6 HSE (2005) Development and validation of the HMRI
safety culture inspection toolkit, RR365, May 2008 www.hse.gov.uk/research/rrhtm/rr365.htm
7 Intn’ Union of Railways (2004) SafeCulture: Method
for Assessing Organisational Safety at Interfaces
8 The Keil Centre Ltd (2003) Managing Safety Culture
in the UK Rail Industry: Report on the Review of
Safety Culture Tools and Methods, RSSB website
9 RSSB (2005) Managing Safety Culture in the UK Rail
Industry: Report on the Review of [23] Safety Culture
Tools and Methods, available via the RSSB website
10 RSSB (2005) Safety Culture, Human Factors Fact
Sheet 1, Rail Safety
11 RSSB Safety Culture Toolkit: www.rssb.info-exchange.
com (as of May 2008)
Understanding Human Factors/June 08
Part 2: Guidance
Culture
How does a high reliability organisation differ from
other organisations?
they spot. Rigid hierarchies are particularly vulnerable
of ‘latent failures’. These are loopholes in the system’s
to error. Errors at higher levels tend to pick up and
defences, barriers and safeguards, such as imperfections
combine with errors at lower levels, making the resulting
in
supervision
and
the
reporting
of
defects,
and
the
1. Preoccupation with failures
problem bigger, harder to understand, and more likely
probable incompleteness of safety procedures, safety
High reliability organisations (HROs) treat any failure,
to escalate. To prevent this deadly
training,
briefings,
certification,
and
whether large or (as is usually the case) small, as a
‘HROs
expect
the
scenario, HROs push decision
hazard
identification.
Many
latent
symptom that something is wrong with the system
failures are discovered only after an
– something that could have severe consequences if
unexpected – they have making down – and around. People
with the most expertise, regardless
accident
has
occurred,
but
this
need
it were to coincide with others at one awful moment.
far
fewer
accidents
than
of their rank, make the decisions
not be the case. We can spot the
HROs encourage the reporting of errors and nearon the front line. This is not
deficiencies
in
normal
operations
that
misses and set out to learn from them. They are wary
you might expect in
simply a case of deferring to the
of what success can bring – things like complacency, the may lead to unexpected events by
complex and dangerous person with the ‘most experience’.
frequently assessing the overall safety
temptation to reduce safety margins, and the drift into
Experience by itself is no guarantee
automatic processing where people stop thinking about health of an organisation.
environments.’
of expertise, since all too often
what they are doing.
people have the same experience over and over again
4. Commitment to resilience
and do little to broaden it.
No system is perfect. HROs know this as well as
2. Reluctance to accept simplification
anyone, and so develop the ability to detect, contain and
HROs take deliberate steps to avoid the process of
bounce back from the inevitable errors. Resilience is a
One of the best examples of an effective HRO is an
simplification that happens when managers focus on a
aircraft carrier, where there are far fewer accidents than
handful of key issues and performance indicators. HROs combination of keeping errors small and improvising
you might expect in such a complex and dangerous
ways
of
keeping
the
system
functioning.
Both
require
know that the world they face is complex, unstable,
environment. During flight operations, when there is a
deep knowledge of the technology, the system, one’s
unknowable and unpredictable. As a result, they have
co-workers,
one’s
self,
and
the
raw
materials.
HROs
put
need for accurate decisions, they tend to be made by
learned to value people who operate in several
the nearest expert – whatever their rank.
a
premium
on
experts:
well-trained
personnel
with
deep
different areas and can develop a wider perspective.
experience and recovery skills. Such experts imagine
They have also learned to be sceptical of received
Adapted from Weick & Sutcliffe (2001), reproduced with permission.
worst-case conditions and mentally
wisdom and they pay attention to
practise their own equivalent of fire
differences of opinion. It is in these ‘HROs push decision
drills.
differences that new organisational
making
down
–
and
problems can be detected – but
5. Deference to expertise
only if the organisation is listening. around. People with the
HROs value not just expertise, but
most
expertise,
regardless
a diversity of expertise. This helps
3. Sensitivity to operations
HROs are very sensitive to
of their seniority, make the them not only to notice more in
complex environments, but also
operations – because they expect
decisions
on
the
front
line.
’
to do more with the complexities
the unexpected – and are aware
Understanding Human Factors/June 08
Page 95
Part 2: Guidance
Culture
Leadership
What is leadership?
Leadership has probably received more attention from
writers on management and organisations than any other
topic. Over its long history the concept of ‘leadership’
has changed dramatically. Originally leaders were seen
as ‘born, not made’. Leadership was something that you
had (probably as a result of being born into a particular
strata of society) or you did not. Nowadays there is much
greater emphasis on leadership as a set of skills that most
people can learn. It is also now a common expectation
that leadership can be exercised not just at the top of
the organisation but at every level. What is important is
whether the right person takes on the leadership function
to meet the demands of the particular situation that has
arisen.
Managers and leaders: what’s the difference?
In their book Leaders Warren Bennis and Burt Nanus
note that many organisations are ‘over-managed and
under-led’. They stress that management and leadership
are both important – but different.
The manager is concerned with accomplishment, taking
charge and having responsibility.
The leader is trying to influence and guide to bring
about some desired future state.
Bennis and Nanus say managers do things right, leaders
do the right thing..
Summarised from: Bennis & Nanus (1985)
display initiative and self-confidence. A key quality is the
‘helicopter factor’: the ability to rise above the details of
a particular situation and see it in relation to the bigger
picture.
There are many hundreds of different definitions of
Leadership entails a contract between the leader and
leadership. Taken literally the term ‘leadership’ implies
followers. Followers are willing to let a leader have power
‘getting others to follow’. You can think of leadership as
as long as the leader continues to deliver. If the leader
a relationship through which one person influences the
fails to deliver, sooner or later the followers will withdraw
behaviour or actions of other people. This leads to the
their support. This could seriously
question of why some people are
‘You can be appointed affect the organisation in terms of
better able to influence others. Some
productivity, turnover and even safety.
leaders rely on their formal position in a manager but you are
the organisation, their ability to reward
not a leader until your What style of leadership works
and punish others. Other leaders gain
best?
authority through their expertise. They appointment is ratified
Leaders display a great variety of
are recognised as having knowledge
in
the
hearts
and
minds
styles. For example, in decision making
that particularly fits them to take on
leaders are entirely dictatorial,
the leadership role in that situation.
of those who work for some
simply telling others what to do.
Still other leaders rely on the force of
you.
’
John
Adair,,
leadership
Some consult the wider team before
their personality, their ability to impose
development
expert
making the decision themselves. Some
their views on others. Leaders need to
Page 96
delegate decisions to trusted subordinates. Others seek
the participation of all the team or staff.
Leaders also vary in terms of where they focus their
efforts. Some concentrate single-mindedly on the task
to be done, expecting staff to do what is necessary to
achieve the task goals. Other leaders devote most of
their efforts to getting the best out of the other people,
trusting that they will get the job done given the right
motivation (page 117) and direction.
As a result of extensive research, the most widely
accepted current view is that there is no one best style. It
all depends on a number of factors, including:
• the position and nature of the leader
• the nature and diversity of the subordinates
• the task to be done
• the organisational norms
• the wider situation.
What should leaders do?
Leaders are oriented towards the future. Perhaps the
key thing that they do is formulate a vision, a picture of
how things could be. This vision needs to be capable
of exciting and motivating others to bring it into reality.
Closely connected with this, leaders communicate,
tirelessly articulating the vision in ways that can be
understood and generate enthusiasm. Communication is
not only a matter of what the leader says but also what
they do. The successful leader should act as a role model,
exemplifying the values of the organisation. Leaders must
Understanding Human Factors/June 08
Part 2: Guidance
Culture
be able to build trust – and not only trust in themselves
as leaders (largely through ensuring their words and
deeds are consistent). They also need to build the
necessary trust and confidence that the workforce will
need to do what is necessary to make the vision come
about.
Leaders must also strive after self-knowledge. They know
their strengths and how to exploit them, and also know
their weaknesses and how to compensate for them.
Closely connected with self-knowledge is something
called emotional intelligence, which means being able to
recognise their own feelings and those of others, and
being able to manage their emotions well.
How can leaders bring about a safety culture?
Nowhere is the role of leadership more important
than in developing an organisation’s safety culture.
Developing and maintaining a safety culture require
leadership to be exercised at all levels throughout the
organisation. However, senior management have a key
role here as they can formulate the vision (eg what
kind of organisation do we want to be?). They can
also send critical messages both through the way they
allocate resources and the example they set. Particularly
important is how much of their own time they are seen
to devote to safety matters.
The HMRI safety culture inspection toolkit includes two
assessment criteria
specifically relating to
leadership. These are
the first two items in
the list below. They
are followed by two
criteria relating to
attitude to blame,
since this attitude can
largely be seen as a
leadership issue.
• Performance vs safety
priority. Safety should
always be prioritised
as more important
than operational
performance
management should be committed to safety and
should demonstrate this by conducting regular safety
tours in all operational areas. Safety tours should
provide the opportunity for all staff to discuss safety
issues with management.
• Awareness and adherence to personal accountabilities.
Personnel should be aware of, understand and adhere
to personal accountabilities.
• Presence of a just culture. Retribution and blame should
not be seen as the purpose of investigations when
things go wrong. Investigation procedures should
clearly distinguish between different degrees of
culpability (eg blameless, system-induced or negligence
induced errors).
Extracted from HSE (2005), reproduced with permission, © Crown copyright
material is reproduced with the permission of the Controller of HMSO and
Queen’s Printer for Scotland.
All of these criteria concern the spoken and unspoken
messages transmitted by organisational leaders to the
wider workforce. It is important here that leaders are
seen to ‘walk the talk’ – to do what they say. There will
be times when hard decisions have to be made between
actions that are revenue earning but potentially risky, and
actions that will cost money but lead to improved safety.
One such decision commonly facing railway leaders is
whether to take a train out of service because of a defect
that may compromise safety. Another example is whether
to ask drivers to work longer than the recommended
12-hour shift when there is a shortage of drivers. And
having made their decision, are they willing to accept
responsibility for their actions when things do go wrong?
• Safety management
leadership. All
Understanding Human Factors/June 08
Source: Railway Safety (2003)
Page 97
Part 2: Guidance
Culture
Good and poor practice in safety leadership
Leadership area
Good practice
Poor practice
Business case
Accountability
Awareness that good H & S management can generate business benefits
Senior managers directly accountable to Board for poor H & S performance
Senior managers have bonus-dependent personal H & S objectives
Director with nominated responsibility for H & S
H & S seen as a source of ‘cost’
H & S seen as a functional responsibility
No H & S objectives set for managers
No responsible Director
Senior managers involved in all accident investigations
Senior managers lead safety briefings and regularly include H & S matters in other briefings
and presentations
Senior managers all commit to receiving regularly-updated H & S training
Senior managers participate in safety audits and raise H & S questions during routine site
visits
Senior managers follow H & S procedures and practices at all times
H & S is a key factor in contractor selection and monitoring
Spending time within H & S functional role is a key part of career development
H & S policy fully integrated into business processes (eg product, plant design)
Key H & S indicators set, monitored and reported regularly to Board
Processes in place to measure the full costs of H & S failures (eg production losses, down
time)
H & S matters are a regular item for discussion at Board meetings
Senior management actively seek staff feedback on H & S issues
Required rate of return for H & S investments is lower than for other investments
H & S performance and investment maintained at times of commercial uncertainty (eg
closure, disposals)
Commitment to ongoing training of staff and contractors at all levels with mechanisms for
sharing learning
No involvement of senior managers
H & S briefings entirely separate and led by functional specialists
Behaviour
Integration
Monitoring and
measurement
Prioritisation
Learning
No senior manager H & S training
H & S matters not raised by senior managers and they do not participate
in audits
Senior managers ignore H & S rules
Contractor H & S performance not considered a business responsibility
H & S function seen as a backwater (eg for those nearing retirement)
H & S matters not part of business process
No regular monitoring or reporting
No data available on costs of failures
No discussion of H & S at Board level
Management processes do not exist for obtaining feedback from staff
H & S investments must meet standard rate of return
H & S performance falls and investment is suspended during commercial
uncertainty
Training is limited to legal requirements. No formal mechanism for sharing
learning
Adapted from: Ernst & Young’s work for HSE (2000), © Crown copyright material is reproduced with the permission of the Controller of HMSO and Queen’s Printer for Scotland.
The choices that leaders make are soon widely known
throughout the organisation and will indicate whether
safety is an issue taken very seriously or whether leaders
just pay ‘lip service’ to the importance of safety.
The diagram on page 97, Creating a safety culture, shows
Page 98
how creating and developing a safety culture involves a
range of safety management activities. This diagram, from
the Railway Safety Good Practice Guide (2003), illustrates
that although leadership is not the same as management,
in reality the two areas are closely inter-connected. You
can find more about these safety management activities
in the section on Management on page 99. Central to
the role of leadership in developing safety management
is the attention senior managers pay to safety. The Panel,
Good and poor practice in safety leadership, defines a range
Understanding Human Factors/June 08
Part 2: Guidance
Culture
of good and poor practices that will all be seen by the
workforce as indicating the value that senior management
places on safety.
Management
Further information on leadership in safety culture
• thousands of staff carrying out many different
occupations
1 Adair J. (2002) Effective Leadership, Pan
2 Bennis W. & Nanus B. (1985) Leaders: Strategies for
Taking Charge, Harper Collins
3 Ernst & Young (2000) Development of a leadership
resource pack, Offshore Technology Report 2000/098
4 Goleman D. (1998) Working with Emotional
Intelligence, Bloomsbury
5 HSE (2005) Development and validation of the HMRI
safety culture inspection toolkit, RR365
www.hse.gov.uk/research/rrhtm/rr365.htm (as of May
2008)
6 Schein E. (1992) Organizational culture and leadership,
2nd Edition, Jossey Bass
7 Roughton J. & Mercurio J. (2002) Developing an
Effective Safety Culture: A Leadership Approach,
Butterworth-Heinemann
Understanding Human Factors/June 08
What does a manager do?
Running a railway involves:
• countless physical and material assets in the form of
trains, rolling stock, infrastructure, stations, signal boxes,
and so forth
• very large financial resources
• vast amounts of information.
The task of a manager
on the railways is
to plan, allocate,
coordinate, integrate
and monitor the
use of all these
resources to
provide a safe and
reliable train service
to many thousands
of passengers.
Although there are
exceptions, most
managers achieve
their results through
other people. The
manager does not
(as a general rule)
drive the train, sell
the tickets, set the
routes, repair the track, and so on. But they ensure that all
these activities happen when they should, in the way that
they should, and at an economic cost. Management can
be seen as the process of transforming a variety of inputs
into a range of outputs under the influence of a number
of shaping factors (see diagram, Management as a process
of transformation).
What makes a good manager?
Managers take on many roles. They allocate resources,
deal with conflicts and disturbances, provide liaison with
the rest of the organisation, represent the organisation in
the wider world, and much else besides. A manager may
sometimes have demonstrate leadership (page 96).
Page 99
Part 2: Guidance
Culture
To carry out all the varied tasks of a manager and to fill
all the diverse managerial roles, the successful manager
must possess many qualities.
First, they must have the right situational knowledge. This
means knowledge of what has happened, is happening
and is going to happen. This knowledge of the situation
must be underpinned by the right professional knowledge
(eg rules, procedures, working practices). Second, they
must have the right management skills. This means being
able to analyse the situation, to solve messy problems,
to make sound judgements, and to take timely and
correct decisions. It also means being able to manage
and influence other people. Lastly the manager needs
the right personal qualities. They must be mentally and
emotionally resilient, must be able to think on their
feet and come up with creative ideas when necessary.
Importantly the manager must be able to reflect on their
own performance and to learn from experience.
The central tasks in which nearly all managers have
to be competent are defined in a management
standards framework. These standards are the basis
for many management qualifications, notably the NVQ
qualifications in management at three levels:
• Level 3 – Supervisory or First Line Management
• Level 4 – Middle Management
• Level 5 – Senior Management.
The standards have recently been overhauled for the
Government by the Management Standards Centre. Since
Sep 05, new courses have been available based on the
results – Panel, What must a good manager be able to do?
Page 100
What must a good manager be able to do?
The new management standards
What must a good manager be able to do?
The new management standards
Training units
Elements
Training units
Elements
A
Manage self and
personal skills
•A1 Manage your own resources
D
Work with people
•D1 Develop productive working relationships
with colleagues
•A2 Manage your own resources and professional
development
•D2 Develop productive working relationships
with colleagues and stakeholders
•A3 Develop your personal networks
B
Provide direction
•B1 Develop and implement operational plans for
your area of responsibility
•D3 Recruit, select and keep colleagues
•B2 Map the environment in which your
organisation operates
•D5 Allocate and check work in your team
•D4 Plan the workforce
•D6 Allocate and monitor the progress and quality
of work in your area of responsibility
•B3 Develop a strategic business plan for your
organisation
•B4 Put the strategic plan into operation
•B5 Provide leadership for your team
•D7 Provide learning opportunities for colleagues
E
Use resources
E1 Manage a budget
E2 Manage finance for your area of responsibility
E3 Obtain additional finance for the organisation
E4 Promote the use of technology within your
organisation
E5 Ensure your own action reduces risks to health
and safety
E6 Ensure health and safety requirements are met
in your areas of responsibility
E7 Ensure an effective organisational approach to
health and safety
F
Achieve results
F1 Manage a project
F2 Manage programme of complementary projects
F3 Manage business processes
F4 Develop and review a framework for marketing
F5 Resolve customer service problems
F6 Monitor customer service problems
F7 Support customer service problems
F8 Work with others to improve customer service
F9 Build your organisation’s understanding of its
market and customers
F10 Develop a customer focused organisation
F11 Manage the achievement of customer
satisfaction
F12 Improve organisational performance
•B6 Provide leadership in your area of
responsibility
•B7 Provide leadership for your organisation
•B8 Ensure compliance with legal, regulatory,
ethical and social requirements
•B9 Develop the culture of your organisation
•B10 Manage risk
•B11 Promote equality of opportunity and
diversity in your area of responsibility
•B12 Promote equality of opportunity and
diversity in your organisation
C
Facilitate change
•C1 Encourage innovation in your team
•C2 Encourage innovation in your area of
responsibility
•C3 Encourage innovation in your organisation
•C4 Lead change
•C5 Plan change
•C6 Implement change
Source: Management Standards Centre (2005) reproduced with permission
Understanding Human Factors/June 08
Part 2: Guidance
Culture
How can managers bring about a safety culture?
Good practice for railway safety management
Key objective
Good practice
Make sure that safety performance in your
area of responsibility is developed and
improved.
•Put strategies and policies for dealing with safety issues into practice.
•Contribute to a positive safety climate.
•Make sure that your team’s performance is consistent with safety strategies and policies.
•Make sure that your team learns from its experiences
Assess and manage safety risks under your
control.
•Assess how safety risks will affect the team.
•Assess people’s behaviour and attitudes to see how they affect safety risks.
•Make sure the actions you chose to control risks are suitable.
Make sure the necessary resources and
support are provided to carry out work
safely.
•Manage working groups effectively.
•Make sure staff and contractors receive the training and development they need.
•Motivate staff and contractors to work safely.
•Make sure any necessary resources are available.
Manage safety in day-to-day activities.
•Make sure that day-to-day activities are carried out safely.
•Put safety measures into practice.
•Make sure that your team meets current safety targets.
Source: Railway Safety (2003)
•Make sure there is a quick and effective response to problems related to safety.
How can you become a better manager?
Good practice in safety management
On the railways, good practice in safety management
Management is a huge topic and it is beyond the scope
begins with adherence to the general safety legislation,
of this Guide to define best practice for all areas of
the Railway Group Standards and specific company
management. However, it’s important to note that
standards. RSSB has defined both objectives and
underpinning all aspects of good management practice
statements of good practice for reviewing and developing
is management development. For the organisation this
the safety performance of managers on the railways.
means formulating a policy and strategy for continuing
They are listed in the Panel, Good practice for railway safety
management development and making available
management. These objectives aid the development of the
the resources necessary for them to be properly
implemented. At the level of the individual manager, good organisation’s safety culture, as discussed in the section on
leadership (page 96). RSSB also provides
practice entails regular review
a companion publication covering
‘Dealing with small
of performance, the setting of
the safety performance of senior
development objectives, and the
things that go wrong
management, as well as software tools
pursuit of these objectives through
to assist in the performance review
both formal and informal means – is the best way to stop
process.
see supervision and appraisal (page
large
things
going
wrong’
68).
Understanding Human Factors/June 08
While it is the task of leaders to set the ‘destination’
and to lead the organisation to that point, it is the task
of managers to use resources to achieve more specific
organisational goals. While leaders spell out the grand
vision of the organisation’s safety culture, managers play a
key part in reinforcing the message about safety culture
through the plans they draw up, the ways they allocate
resources, and the manner in which they respond to
safety issues and incidents.
One of the less obvious ways that a manager can help
develop a strong safety culture is through developing
a learning culture (one of the HMRI safety culture
inspection criteria – see Panel on page 94, What
should you assess?). A learning culture is a necessary
accompaniment to a safety culture. Without the ability to
learn, especially to learn from things that go wrong, no
organisation can achieve a high level of safety. Lord Cullen
pointed out the need to learn from:
• previous accidents
• near misses
• the analysis of information regarding non-compliance
• analysis of behaviour leading to unsafe acts
• incidents in other related industries
Going back to our earlier discussion of High Reliability
Organisations (HROs) (see Panel on page 95, How does
a high reliability organisation differ from other organisations?),
remember that the first characteristic of an HRO is a
pre-occupation with failure. No incident where something
Page 101
Part 2: Guidance
Culture
goes wrong is too insignificant to be observed, recorded,
thought about and learned from. This may seem a very
negative way to look at working life, but experience has
shown that detecting and dealing with the small things
that go wrong is the best way to stop large things going
wrong.
This approach not only prevents problems escalating,
but it also cuts down on the number of organisational
‘holes’ that might otherwise line up as part of an accident
causality chain (see Why do accidents happen? page 18).
One of the tasks of a manager on the railways is to
ensure that there are proper mechanisms set up and in
use that support this learning process. Such a process
must be founded on a system for gathering information
on what goes wrong, as well as what goes right. The
rail industry does have a national confidential incident
reporting scheme (CIRAS), but there is likely to be
a need for more local reporting initiatives too. In the
event that these schemes are not designed to maintain
confidentiality, remember that they will only work if there
is a no-blame culture, as discussed in the section on
leadership, page 96.
Similarly, the rail industry has well-established methods for
conducting formal inquiries into incidents. However, these
are designed for when something has gone sufficiently
wrong for there to have been damage or injury, or at
least a serious breach of the rules. An organisation with
strong safety and learning cultures will also learn a great
deal from much more minor events that were not in
themselves serious but may point to potential areas of
vulnerability.
Page 102
But simply collecting information is not enough. It needs
to be analysed in the right way, so that the right action
can be taken. Plotting trends may be a good way of
seeing that a situation is deteriorating, thus enabling
timely action to be taken before a major incident arises.
A good example here is provided by the RSSB’s use of
the methods of statistical significance testing to analyse
monthly SPAD data. This is a powerful technique not
only for obtaining an early warning of potential problems,
but also for identifying conditions under which risks
are especially high, and for assessing the impact of any
remedial actions that may have been taken.
By encouraging staff participation in gathering and
discussing information on incidents and concerns, even
the most minor, and in showing that the analysis of this
information leads to positive actions, you can foster a
climate in which everyone takes safety seriously and acts
accordingly.
Further information on management
Any bookshop can provide a huge range of books on
various aspects of management. Many of these are rather
gimmicky, offering instant solutions to all the problems of
management. The first three of the following books are
views of management by well-respected writers and the
second three are good general accounts of management.
The last item in the list is a short book that is an easy to
read introduction to the methods of Statistical Process
Control. Only the most basic knowledge of arithmetic is
required.
1 Adair J. (2004) The John Adair Handbook of
Management and Leadership, Thorogood
2 Armstrong M. & Stephens T. (2005) A Handbook of
Management and Leadership, Kogan Page
3 Crainer S. & Dearlove D. (Eds), (2004) Financial Times
Handbook of Management, Financial Times Prentice
Hall
4 Drucker P. (1999) Practice of Management,
Butterworth Heinemann
5 Handy C. (1999) Inside Organisations – 21 Ideas for
Managers, Penguin
6 Management Standards Centre (2005)
www.management-standards.org (as of May 2008)
7 Mullins L. (1999) Management and Organisational
Behaviour, 5th Edition, Financial Times/Pitman
Publishing
8 Railway Safety (2003) Good Practice Guides (2):
Reviewing and developing the safety performance of
managers
9 Schermerhorn J.R. Jr. (2005) Management, 8th Edition,
John Wiley
10 Wheeler D.J. (1993) Understanding variation – The
key to managing chaos. SPC Press, Knoxville, Tennessee
Understanding Human Factors/June 08
Part 2: Guidance
Culture
Teamworking
What is a team?
Definitions
Team
A group of at least two people, in which each person
depends on the work of the other(s) to achieve their
own objectives.
Teamwork
How members of a team work responsively with
each other to ensure that all the team’s objectives are
achieved.
Performance and safety on the railways depend on
many kinds of teamwork. Usually when we think of a
team we think of something like a football team. Here,
team members spend extended periods together,
practising and working until their efforts can be smoothly
coordinated to achieve the desired results. They know
each other as individuals and ideally learn to trust each
other. At the very least they know each other’s strengths
phone). Furthermore, their working together does not
and weaknesses, and can take these into account when
last over an extended period of time. Its duration is
working together. Signallers working together in a large
probably measured in minutes. So unlike some of the
signal box or signalling centre, the
other examples of teams on the
staff of a major station, the crew on
‘Teams on the railways railways, these are highly transient,
a long-distance express train, or an
two-person teams. They probably do
may form for only a
experienced gang of track workers are
not have any personal knowledge of
all examples of this kind of team.
few minutes and its
each other. Nor can they practise their
teamwork skills. Team building, which is
members may never
But there are other kinds of team that
discussed in the section on supervision
play a vital role in the safe and effective meet each other…’
and appraisal (page 68), cannot
running of the railways and which do
be used for these kinds of team.
not have the conventional attributes of a team.
Nonetheless, you can think of these driver/signaller and
Consider two examples. The first is a signaller in their
signal box and the driver of a train passing through this
signaller’s section. The second is the COSS (Controller
of Site Safety) and the ECO (Electrical Control
Office) when the COSS requires the electricity to be
disconnected from the lines so that engineering work can
take place. In both cases, the two interacting workers are
not located in the same place. Indeed, they may never
ever actually meet each other. All their work is via some
form of communication channel (for example, Signal Post
Telephone (SPT), Cab Secure Radio (CSR) or mobile
Understanding Human Factors/June 08
COSS/ECO pairs as teams according to the definition
given in the Panel, Definitions. These definitions do not
assume that the team is either co-located or long-lasting.
What makes a good team?
Teamwork skills
What a good team needs first and
foremost is for all its members to
possess high-quality teamwork skills.
Teamwork skills are different from
‘taskwork’ skills, which are those that
enable a train driver, say, to drive their train, or a signaller
to regulate train movements. These are usually well taught
as part of a person’s occupational training. But in addition,
staff on the railways need to be able to work effectively
with others. Research has identified what these key skills
are. They include anticipation, cooperation, and challenging
and backing up each other. However, it has also shown
that these sorts of skills often do not form a significant
part of formal training but are left to be picked up on the
job.
Good organisational practice ensures that any members
of staff on the railways, especially those involved in safety
critical tasks, acquire these skills as part of their initial
training. They should also have ample opportunities
to practise and develop these skills throughout their
career. Individuals whose job requires them to work
briefly, but critically, with a range of other individuals
over the working day need to be equipped with the
fundamental skills that enable them to work effectively
and immediately within teams that form and disband in a
few minutes.
An organisational context for teamwork
As well as having reasonable expectations that other
staff will possess a satisfactory level of the necessary
teamwork skills, a good team needs
a sound organisational context for
‘Effective teamwork
teamwork. Having teamwork skills
requires both team
means that a person knows how to
behave as a team member, but does
skills and a cooperative
not mean that they know what to
organisational context’
do. Good teamwork is only possible
when team members believe they
Page 103
Part 2: Guidance
Culture
can rely – within limits – on what other members of staff
will do in any given situation. Excessive uncertainty about
what the other person may do is highly damaging to
effective teamwork.
An important function of the Rule Book is to provide
an industry-wide framework which dictates in many
situations – but by no means all – what people should
do. The Rule Book reduces much of the uncertainty
when people are working together, but more is needed.
This is where the organisation’s culture comes in, for it
determines many of the ‘softer’ aspects of how people
work with each other (see change on page 109).
The values, norms, customs and social practices that
characterise the railway industry as a whole are key here.
But there is a particular issue nowadays, for within the
overall railway culture there are many company-specific
sub-cultures. The existence of these sub-cultures can
pose a risk to effective teamwork. It is the responsibility
of leaders at both the industry and company levels to
ensure that these sub-cultures all promote rather than
impede teamwork.
What is teamwork good practice?
Good practice for teamwork exists at two levels.
At one level, good practice defines what individual
team members should do to be effective members
of successful teams. But at a more fundamental level,
good practice defines what companies or organisations
should do to ensure that their staff are able to behave in
accordance with individual good practice. Organisations
need to create an environment in which teamwork can
flourish. If they do, they can have a profound effect on
the ability of team members to work with each other.
Page 104
Good practice in teamworking
Good organisational practice
Good team member practice
Personnel in safety critical roles should be trained in how to carry out
safety critical communications
Personnel should always communicate safety critical information to those
who need to know and confirm the message has been received and
correctly understood
All initial training for safety critical occupations should cover basic
teamwork knowledge and skills
Personnel in safety critical roles should communicate at a time and in a
manner that will be most helpful to the others and minimises the risks of
misunderstandings
All safety critical personnel should have regular opportunities to practise
teamwork skills, especially in abnormal or emergency situations
All personnel in safety critical roles should seek clarification of
communications from others in the event of ambiguity or uncertainty
All personnel should have regular opportunities to spend time with other
personnel with whom they perform safety critical functions, including time
observing them at work
Personnel in safety critical roles should monitor others’ situations and the
demands they are facing
Competence in teamwork should form an element in all recruitment and
assessment for personnel in safety critical roles
Personnel in safety critical roles should allow for the effect of one’s actions
on others in the choice and timing of the action
Personnel performing safety critical functions should be provided with an
infrastructure that enables them to build and maintain shared awareness of
each other and their working situations
Personnel in safety critical roles should challenge the decisions or actions of
others that may be unsafe
Supervisors, shift managers, and others responsible for the work of teams
performing safety critical functions should be selected for and trained in
leadership
Personnel in safety critical roles should anticipate the needs or problems of
others and take timely action
Representatives of all groups engaged in safety critical functions should
meet regularly to review teamworking practices, identify improvements and
introduce these improvements to their colleagues
Personnel in safety critical roles should provide help or support to others
who need assistance
Any proposed changes to rules, procedures, working practices, systems and
workplaces should be evaluated for their potential impact on teamwork
- throughout the company
Personnel in safety critical roles should provide feedback about others’
performance and accept feedback about their own performance
All companies employing personnel involved in safety critical teams should
have a policy, supported by systems and practices, for how they promote
teamwork – especially with other companies
Source: Gregory & Shanahan for RSSB (2004)
Personnel in safety critical roles should strive to develop an open and
positive team climate amongst all those with whom they interact
Unfortunately, the opposite is also true. See Panel, Good
practice in teamworking.
How does teamwork affect safety culture?
It is clear that good quality teamwork is a necessary
ingredient in safety. In particular, it is the ability of
members of effective teams to monitor and support each
other that enables many human errors to be avoided,
or at least detected and dealt with. More generally,
teamwork has a two-way relationship with safety culture.
On the one hand, a strong safety culture will naturally
express itself in terms of high quality teamwork –
members of teams will be motivated to support each
other and will know how to do so. On the other hand,
Understanding Human Factors/June 08
Part 2: Guidance
Culture
Further information on teamworking
teams are important vehicles for transmitting the values
of the organisation’s safety culture. How team members
behave towards each other, and the expectations they
have, are ways in which safety values are demonstrated
and upheld. Furthermore, a strong safety culture demands
the ability to respond flexibly to difficulties and threats.
Teams contribute significantly to that ability. Good
teamwork depends on good team leadership (page 96)
and good communications (page 106).
1 Gregory D. & Shanahan P. (2004) Teamworking best
practice in the railway industry: The Journey Guide,
Gregory Harland Ltd, for RSSB, Euston (available via
the RSSB website)
2 Katzenbach J.R. & Smith D.K. (1993) The Wisdom of
Teams McGraw Hill
3 Robbins H. & Finley M. (2000) Why Teams don’t Work,
Tenere
How can you improve teamworking?
In 2004 the RSSB published guidance aimed specifically at
showing railway organisations how they can use the best
practice outlined in this Guide to improve teamworking
(see Teamworking Improvement Process). This teamworking
guidance (see Further information) describes a process
that can be run in-house and takes your organisation
through three stages. In the first stage, you put together
a small steering group and use a simple questionnaire to
diagnose the nature of any teamworking difficulties at a
workshop. In the second stage the process helps you to
identify and prioritise the interventions that will be most
cost-effective for your organisation to undertake, given
the nature of the team working diagnosed. This second
stage can take place at a second workshop soon after
the first – and often on the same day. The final stage
takes place some months later and aims at measuring the
new state of teamworking that has been created by your
teamworking improvement interventions.
The diagram illustrates the overall direction of the team
improvement process.
Understanding Human Factors/June 08
4 West M. (1994) Effective Teamwork, British
Psychological Society
Source: Gregory & Shanahan, for RSSB (2004)
Page 105
Part 2: Guidance
Culture
Communication
Why is communication so important?
Good communications are vital to organisational
effectiveness, and yet nearly every organisation reports
communications as a problem area. For an industry
such as the railways, where much communication is
safety critical, this is an area that has to be got right.
Recent research by RSSB indicates that around a third
of incidents sufficiently serious to require a formal
investigation are at least partly caused by communication
failures of some kind. For this reason we focus here upon
safety critical communications.
The railway industry has long recognised the importance
of good communications to safety. Much effort has been
devoted to researching communications issues and to
developing rules and codes of practice for safety critical
communication. In terms of the formal requirements the
three key documents are:
• Railway Group Standard, GE/RT8046, Safety
Communications, dated October 2002
• Railway Safety Approved Code of Practice, Safety
Communications, GE/RC8456 dated October 2002
• Rule Book, GE/RT8000, Module G1, General Safety
Responsibilities, Section 11, Issue 1, June 2003
In addition, a new Rule Book module on Communications
was due in Dec 2005.
In recognition of the importance of safety critical
communications, the railway industry has set up the
SCCFG (Safety Critical Communications Focus Group).
Page 106
This group has as its first objective ‘to raise the profile of
safety critical communications within the industry and to
encourage cross-industry efforts to improve its quality’. Its
principal aim is ‘to encourage a culture of professionalism in
safety critical communications – a culture where everyone
– managers, supervisors and front line staff - appreciates
the role communication plays in reducing safety risks and
improving performance, and behaves accordingly’.
Source: Rail Safety Critical Communications website
Why is communication so difficult?
Communication is often problematic because so many
things can go wrong. When, for example, two people
in different places try to communicate over some
communication channel, such as a mobile phone or radio:
• People forget. Especially when under pressure from a
high workload (page 125) or even stress (page 120),
people forget to tell other people things they should
know. They may not think to call them at all. Even
when talking to the other person they may forget to
say things they should, or they may forget exactly what
it was they wanted to say.
‘A third of all railway incidents
sufficiently serious to require a
formal investigation are at least
partly caused by communication
failures of some kind’
may be noisy. One of the people may be in a noisy
place. Accents or dialects may make it difficult for one
person to understand the other. One or both may fail
to use the phonetic alphabet. The person listening is
distracted and does not attend carefully to what the
other is saying.
Even when two people manage to make contact and can
hear each other properly, many things can still go wrong:
• Slips of the tongue. One person inadvertently says
something different from what they intended.
• Use of jargon. One person uses words or terminology
that the other person is unfamiliar with so that they
either do not understand at all or interpret wrongly.
• People cannot get into contact. The other person may
• Wrong assumptions. One person may not realise that
be engaged in some other conversation. One person
they are talking to somebody other than the person
may not have the other’s phone number. One person
they intended to talk to. Or the person
may be in a position where
may actually be somewhere different
their mobile phone cannot ‘Even when two people
from where the other thinks they are.
receive a signal. Their
manage
to
make
contact
One person may assume the other is
mobile phone battery may
talking about X when they are actually
and can hear each other
be out of charge.
talking about Y – and neither of them pick
properly, many things can this up.
• People mishear. The
communication channel
still go wrong’
Understanding Human Factors/June 08
Part 2: Guidance
Culture
• The power of expectations. People have a strong
tendency to hear what they expect to hear.
• Lack of clarity. One person may describe a problem, a
situation or an intended action in a way that the other
misunderstands what they are being told (see Panel,
… know what I mean?)
• Failure to do a ‘sanity check’. One person listens
uncritically (‘brain in neutral’) and does not consider
whether what they are being told makes sense.
Given all the things that can go wrong, the wonder is
not that communications go wrong so often, but that
they actually succeed most of the time. The railway rules
and procedures for safety critical communications are
designed to minimise the risk of many of the problems
above. However, it is worth noting that having to speak
in a highly formalised way is itself more difficult than
talking in plain language. Having to think about rules
and procedures is itself an additional mental load that
can contribute to communication errors – unless they
become ‘automatic’ through extensive practice.
How do communications help with safety
culture?
So far we have considered the immediate contribution
that good communications make to safety in the context
of safety critical communications. And of course goodquality communications are the foundation of effective
teamwork. But when we consider an organisation’s safety
culture we need to look more widely at communications.
The case for good communication in developing a safety
culture can be traced back to the Ladbroke Grove
Understanding Human Factors/June 08
… know what I mean?
‘The protection has been placed on the station side of one
six six seven points.’
inquiry. Two-way communication is also one of the five
key indicators of safety culture examined by HSE’s HMRI
toolkit (see Further information). This toolkit for inspectors
picks out three aspects of communication:
On the face of it, this might seem like a pretty
reasonable statement for one railway professional
to communicate to another. It comes from an actual
recorded conversation between a COSS and a signaller.
But, during a lengthy discussion about the ‘station side’
of a set of points, it became clear that they were in
fact talking about different stations on either side of
the points. With personnel working on one side of the
set of points while the signaller thought they were on
the other side, a major train accident or fatalities could
have occurred.
• Promotion and awareness of safety culture. All personnel
should be aware of and understand safety goals,
targets and issues. There should also be visible
efforts by senior management to communicate their
commitment to developing a positive safety culture.
The solution to a situation like this is to be as precise
as possible. That doesn’t mean you should regurgitate
a section of the Rule Book, just in case the person has
forgotten their training. But do:
• give names of locations and people.
• give the status of an event: is it going to happen (if
so is there a specified time), has it happened or is it
happening now?
• use the specific specialist terms when appropriate
(eg ‘this is an emergency call’, ‘there has been a train
accident’)
• be aware that not everybody thinks and speaks the
same way as you. Your reference to ‘inter’ might be
understood as ‘into’ by another person, rather than
‘Birmingham International Station’ as you intended.
Source: Rail Safety Critical Communications website
• Safety concern reporting. There should be clear and easy
to follow procedures to report safety concerns. The
reporting system or process should be accessible to all.
• Discussion and awareness of safety issues. There should
be multiple channels for communication about safety.
Extracted from HSE (2005), reproduced with permission, © Crown copyright
material is reproduced with the permission of the Controller of HMSO and
Queen’s Printer for Scotland.
The HMRI toolkit stresses communications should be
two way, from staff in the front line, through supervisors
and middle managers to those at the very top of the
organisation. And of course back down the organisation
in the other direction. It is also worth remembering
that communications is two-way in the sense that
communication only happens if there is both speaking
and listening:
Good communications to support a safety culture will
use many different channels, formal and informal, and a
variety of communication media, such as briefings, notices,
videos, etc. There is a common tendency for safety
concerns to be expressed verbally by front line staff to
their immediate managers. These concerns also need
Page 107
Part 2: Guidance
Culture
to be put in writing, and if necessary the organisation
should establish a central database for gathering all these
reports. It is vital to identify problems in order to take
pre-emptive action. (See also the discussion of learning
culture in the section on Management on page 99.)
Good practice in communications for safety critical staff
Start on the
right note
•identify yourself, your job title and your location
•ensure you identify the other person
How can you improve communications?
The SCCFG website (www.rail-scc.co.uk, as of May 2008)
has an excellent summary of good practice for safety
critical communications by front-line staff (see Panel,
Good practice in communications for safety-critical staff).
The website also gives you The Safety Communications
Good Practice Handbook produced for RSSB in 2004 by
Risk Solutions. This handbook covers all aspects of safety
communications from the perspective of senior managers
and directors, safety and line managers, and briefers.
How can you manage communications?
In addition to rules and guidance on actually
communicating safety critical information, the rail industry
has formulated good practice on what Railway Group
members must do in order to ensure that safety critical
communications are maintained at the highest standard
possible. GE/RT8046 and GE/RC8456 specify what the
management of organisations in the railway industry must
do in terms of:
•be clear about the purpose of the call.
The ABC
principle:
Accurate
Brief
Clear
• establishing a safety critical communications
competence management system
Page 108
The principles of safety critical communications are simple: keep the conversation accurate, brief, and clear.
•Speak slower than normal.
•Use clear sentences and make sure you use the standard phrases.
•Avoid technical and regional jargon – not everyone is familiar with it.
•Spell words that are unfamiliar or difficult to pronounce. Use the phonetic alphabet.
•Read back to confirm understanding.
•Say numbers individually. Read back to confirm understanding. Remember how easy it is to get numbers and codes wrong by a slip of
the tongue.
•Take time to reach understanding.
•If the caller speaks in a dialect or accent which is not familiar to you, what they say may take some repeating.
Lead
responsibility
Lead responsibility is an important element in safety critical communications. It means that one person guides the conversation and
takes charge of the outcome. That person has the authority and duty to show the way forward and direct the action. Because it is so
important, lead responsibility is specifically assigned to different work roles, and varies according to who you are communicating with.
Listening skills
It may be obvious that listening carefully is as critical as speaking clearly. But how do you know if someone is listening? Asking them to
repeat the message is a simple way to check it. The protocols give you several ways of doing so. Asking for a read back and repeating
messages is always a good practice, but even more important in situations when there is a risk of inattention and lack of focus. On the
railway these include:
•long-term degraded working conditions – when exception becomes normal
•monotony of work
•tiredness
•stress
•intense activity
• defining safety critical communication requirements
• recruiting and selecting staff to undertake safety critical
communications
You have probably experienced in everyday life that when a conversation starts on the right note, it usually carries on that way, too.
The same applies in safety critical communications. Research from the railway shows that when conversations start well, as per the
protocols, they tend to continue that way. When you are the first to use the protocols, you help to set the professional tone. When
beginning a communication, you must:
•adverse conditions – wanting to work quickly rather than safely.
Confirm
understanding
The ultimate purpose of the protocols is to ensure that clear understanding is reached about all critical details in safety critical
operations. No action should be taken without confirming understanding.
Source: Rail Safety Critical Communications website
Understanding Human Factors/June 08
Part 2: Guidance
Culture
• training for safety critical communications
Change
• monitoring safety critical communications performance
What drives change?
• gathering safety critical communication data
If you are responsible in any way for safety critical
communications performed by railway staff, you
should ensure you are familiar with these management
requirements and make the appropriate arrangements.
Further information about communications
1 Clark H. (1996) Using Language, Cambridge Uni Press
2 Cushing S. (1994) Fatal Words: Communication
Clashes and Aircraft Crashes, Univ. of Chicago Press
3 Dietrich R. (Ed) (2003) Communication in High Risk
Environments, Helmut Buske Verlag
4 HSE (2005) Development and validation of the HMRI
safety culture inspection toolkit, RR365
www.hse.gov.uk/research/rrhtm/rr365.htm (as of May
2008)
5 Safety Critical Communication Focus Group
www.rail-scc.co.uk (as of May 2008): several
publications including The Safety Communications
Good Practice Handbook produced for the Rail Safety
and Standards Board by Risk Solutions (2004)
It has become a truism that the only constant in working
life nowadays is change. But it is certainly the case that
the railway industry has seen its full share of change over
the last decade or so. Probably the major driver of change
over this time has been a series of Government decisions,
notably the decision to privatise the railways in the early
1990s. These changes in the structure of the railways
have been accompanied by increased exposure to new
financial and business pressures. Other drivers of change
have been the influence of initiatives at the European
level (eg the EU Working Time Directive) and the growth
in the number of train passengers.
What are the obstacles to change?
In any organisation there is a continual process of
evolutionary change. This involves numerous small
changes triggered by any of a multitude of factors.
Normally the organisation will adapt to these changes
without major disruption. But usually when people talk
about the problems of organisational change they have in
mind larger programmes of planned, intentional change.
Such programmes are notoriously
‘Nothing is permanent difficult. Many of these problems are
practical in nature, but here we focus
except change.’
on what most writers on organisational
Heraclitus 6th Century BC,
change single out as the greatest issue,
Greek philosopher
namely the human factor in change.
As it has throughout its history,
the railway industry has to remain
responsive to changes in the available
technology. For example, the
availability of cheap, reliable mobile
phones has had a strong impact on
many areas of railway working. This
impact will only increase as new programmes such as
GSM-R and ERTMS change fundamentally many aspects
of how the railways work. Specific changes in working
practices have also been driven by inquiries into major
accidents, such as the Hidden Inquiry into the 1988
Clapham accident and the Cullen Inquiry into the 1999
Ladbroke Grove accident.
There have also been changes that have affected the
workforce in less obvious ways. Over the years skills have
varied as education and training practices have developed.
The composition of the workforce has changed under
Understanding Human Factors/June 08
the impact of factors such as immigration and the
increasing acceptance of women in many occupations
that were traditionally ‘male’. Social attitudes have also
changed. For example, few people these days expect to
have a career for life in the way that earlier generations
did. All of these changes are reflected in change in the
way that the railways are staffed and operated.
The human factor in change normally takes the form of
resistance to change. Fear of the unknown is a leading
problem. Staff fear that they may lose out financially or
in terms of loss of power or influence. They may have
fears that they may lose their jobs. Where the change
involves new technology or working practices people
may have concerns about safety. Even when fear is not
a large concern, change always involves disruption and
inconvenience. A lot of working life is about habits.
Having to give up old habits and learn new ones is usually
unwelcome.
Page 109
Part 2: Guidance
Culture
More generally, organisational culture can play a major
role in holding back the process of change. Many will view
attempts to change the culture as putting at risk the very
things that have made the organisation successful in the
past – the shared values, beliefs and assumptions that
have evolved over the years. Yet it may be essential to
change the organisation and culture if the organisation is
to remain viable. This is likely to be a challenging process
since much of culture is deeply embedded in people at
a subconscious level, and it is remarkably difficult even
to identify what needs to change, let alone to actually
change it.
How can change be managed?
Change is driven by many different factors, which means
that change management programmes can take many
different forms. In keeping with the themes of this Guide,
we look here at three highly inter-connected aspects of
managing change: technological change, change with safety
implications and cultural change.
Introducing new technology
The railway industry is of course technology-intensive.
There is a continuous process of replacing older
technologies with new ones. This has always been the
case with the technologies connected with trains, track
and infrastructure, but more recently the rapid advance of
information and communication technologies has resulted
in radical changes in management and administration.
The introduction of new technology into organisations
has not always been a story of success. New technology
has often triggered periods of disruption and even
conflict, especially where people have felt threatened by
Page 110
the technology. Projects to introduce new technology
have often been abandoned or the new technologies
have soon fallen into disuse. The financial and other costs
have been enormous.
Studies of why so many attempts at introducing new
technology have failed have consistently pointed to a
major cause: the failure to properly consider the human
and social working context into which the technology
is introduced. Managers have often focused on the
operational or business benefits expected of the new
technology, and designers have been totally absorbed
by the technical challenges it poses. The fact that
the technology may disrupt or make obsolete wellestablished working practices, relationships and senses of
professional identity is frequently ignored.
To avoid these difficulties, an approach to the introduction
and development of systems, called socio-technical
design, has been developed over the last 50 years. The
essence of this approach is the recognition that the vast
majority of working situations involve teams or groups
of people working with each other (the social system)
and with complex combinations and arrangements of
technology (the technical system). Both the social and
technical systems have to develop in step with each other.
The aim is to find a solution in which business-oriented
and human objectives can be met to the reasonable
satisfaction of both types of system. A process for
achieving this is outlined in the diagram on this page,
Socio-technical systems design. You can view this process
as providing a wider context for the user-centred design
process (see section on user-centred design on page 25).
Understanding Human Factors/June 08
Part 2: Guidance
Culture
Change with safety implications
Getting organised checklist
Don’t make too many simultaneous changes, resulting
in inadequate attention to some or all.
‘Organisational change is often an opportunity to improve
health and safety, for example though reappraisal of
safeguards or clarification of personal accountabilities.
However, HSE’s experience is that in many instances
organisational changes are not analysed and controlled
as thoroughly as plant changes, resulting in reduced
defences against major accidents, sometimes with fatal
consequences. This is because, unlike management of
plant change, impacts of organisational change are less
well understood, and there is a lack of robust, generally
accepted approaches to ensuring safety.’
HSE (2003)
Don’t delay or defer safety issues compared to other
aspects considered more pressing, because:
• they are seen as a side issue
• they are delegated to people with inadequate
influence
• they are not considered early enough in the change
process
• inadequate time or resource is allocated to their
assessment
Organisational change can, of course, include the
introduction of new technology. It can also include
outsourcing, mergers, introduction of ‘self managed’
teams, and so on. To help overcome the problem of not
assessing organisational changes sufficiently for their safety
implications, the HSE has developed the change process
shown in the diagram, Managing organisational change.
There are three major steps, each of which involves a
number of tasks.
• teams making decisions are too inward looking
• there is lack of objectivity
• objectives are passive, maintaining rather than
improving standards
• appropriate management controls are missing
Source: HSE (2003), reproduced with permission. © Crown copyright material
is reproduced with the permission of the Controller of HMSO and Queen’s
Printer for Scotland.
The HSE information sheet (CHIS7) not only gives a
detailed description of the process but also a number
of valuable checklists. The HSE overview checklist is
reproduced in the Panel, Getting organised checklist.
Can safety culture be changed?
Within the context of this Guide, the organisational
change that is of most relevance is changing the safety
culture. We have already considered safety culture in
Understanding Human Factors/June 08
Source: HSE (2003), reproduced with permission. © Crown copyright material is
reproduced with the permission of the Controller of HMSO and Queen’s Printer
for Scotland.
relation to leadership, management, teamwork and
communications. Introducing any of the good practice
discussed in these sections will begin to bring about
changes in your organisation’s safety culture. However,
there will be occasions when you seek to bring about
more comprehensive changes in your organisation’s safety
culture.
Page 111
Part 2: Guidance
Culture
You can see changing safety culture as a specific example
Stage 1 is a dependent safety culture. Here the emphasis
of changing organisational culture
is on management and supervisory
in general. An important point here ‘Culture can only be
control, with extensive use of
is that culture cannot be changed
to enforce safety measures.
changed indirectly – and discipline
directly. It can only be changed
There is a heavy reliance on written
always takes a long time’ safety rules and procedures. Safety
indirectly, and this will probably
take a long time. Culture can slowly
performance is dependent on the
change through the operation of a
level of management commitment
range of organisational ‘levers’, such
to enforcing rules and procedures.
as leader behaviour, changing rules
Safety performance improvement will
and rewards, and visible things such as buildings, names,
reach an upper limit with this type of culture – because
and so on. What is clear is that because an organisation’s
no matter how committed management are, it is not
culture is deeply embedded in the minds and identities
possible to be everywhere and observe all operations.
of the members of an organisation, any change in
organisational culture usually only happens when a crisis
Stage 2 is an independent safety culture. Here, the focus
confronts the organisation and its staff, (see Panel, The key
ingredients of culture change).
is on personal commitment to, and responsibility for,
safety. This involves all employees in developing their
own personal safety standards and demonstrating their
commitment by adhering to these standards. While
there are still safety rules and procedures, employees
look after their own safety and make active choices to
keep themselves safe. In an independent safety culture,
the focus on individual responsibility for safety may be
indicated by statements such as ‘everybody is their own
safety officer’.
Stage 3 is an interdependent safety culture. Here,
team commitment to safety is the dominant
factor. This type of culture is manifested by
workers having a sense of responsibility for safety
The diagram, The levers of culture change, illustrates the
mechanisms involved in bringing about culture change.
Change management research in the off-shore industry
has shown that an organisation’s safety culture typically
goes through three stages of development as it changes
and matures.
The key ingredients of culture change
• Typically triggered by a perception of crisis
• Initiated and shaped by strong leaders
• Consolidated by perceived success
• Requires extensive re-learning and re-education
Adapted from Scott et al (2003), reproduced with permission
Page 112
Adapted from Briault (1994), reproduced with permission
Understanding Human Factors/June 08
Part 2: Guidance
Culture
‘Culture is always influenced
by management actions
– for better or worse’
beyond their own work and by caring for the safety
of others. Employees share a common belief in the
importance of safety. The movement towards an
‘interdependent’ culture is difficult, as it relies on
more than personal commitment; it requires shared
perceptions, attitudes and beliefs. Employees need to be
willing to help others to adopt this belief system – not by
sanction but by persuasion.
Adapted from: Fleming & Lardner (1999), reproduced with permission from
The Chemical Engineer
What are the principles of change management?
We have looked at ways of introducing new technology,
improving safety or changing culture. These each have
their different demands and requirements. But research
from many areas has indicated that there is a common
set of principles that you can regard as good practice
in managing change (see Panel, The principles of change
management).
The principles of change management
• An important priority is to create an environment of
trust and shared commitment, and to involve staff in
decisions and actions which affect them.
• There should be full and genuine participation of all
staff concerned as early as possible, preferably well
before the introduction of any change.
Further information on organisational change
The first of these references is a short publication highly
relevant to the theme of this guide. The others are all
books that present important views on the process of
organisational change in general.
1 Briault S. (1994) What is integrated learning? Training
Officer, May, Vol. 30, No. 4
• Team management, a cooperative spirit among
staff and unions and a genuine feeling of shared
involvement will help create a greater willingness to
accept change.
2 HSE (2003) Organizational change and major accident
hazards, Chemical Information Sheet CHIS7
• As part of the pre-planning for any change,
there should be a ‘personnel management action
programme’ that is carefully designed to prepare
staff for the change and to minimise any negative
consequences of the change.
4 Mumford E. & Beekman G.J. (1994) Tools for change
and progress, CSG Publications
• The introduction of incentive payment schemes
– based on an equitable allocation of the savings
that may result from the changes and more efficient
methods of working – may help in motivating staff.
3 Kanter R.M. (1983) The Change Masters, Unwin
Paperbacks
5 Scott, T, Mannion, R, Davies, & H, Marshall, M. (2003)
Healthcare performance and organisational culture,
Radcliffe Medical Press
6 Senge P., Kleiner A., Roberts C., Ross R., Roth G.,
& Smith B. (1999) The Dance of Change, Nicholas
Brealey
• Changes to the work organisation must maintain the
balance of the socio-technical system.
• Careful attention should be given to job design,
methods of work organisation, the development
of cohesive groups, and relationships between the
nature and content of jobs and their task functions.
Adapted from Mullins (1999), reproduced with permission
Understanding Human Factors/June 08
Page 113
understanding
© Four by Three 2003
Conditions
human factors
understanding
Conditions
human factors
© Four by Three 2003
Part 2: Guidance
Conditions
Conditions
This section is concerned with an important set of human
factors which influence the conditions in which people
work. The diagram, Focus on conditions, shows the four
sub-areas that this section focuses on (in the middle red
ring), and identifies the main human factors questions that
this Guide answers (in the outer grey ring). At the end
Focus on conditions
of each section, you will find a list of sources of further
information that will provide more detail.
In this section we look at morale and motivation, how
they are related, and what can be done to improve them.
Morale and motivation
What are morale and motivation?
In the early 1990s, during the run-up to privatisation,
one of the authors of this Guide worked with a group
of British Rail S&T supervisors and team leaders on
a management training course. It was a time of great
uncertainty, with most of the group unsure what would
happen to their jobs. They also had a lot
of extra tasks to do to prepare for the
privatisation, such as documenting the
quality system, even though they did
not know if they would benefit
personally from this work. Not
surprisingly, morale was at rock
bottom. The group’s motivation
to learn about management
when they did not know
whether they would have
a job in a year’s time was
non-existent. They learned
little from the course.
Morale and motivation are two concepts in managing
people that are closely related and important, but hard
to define (see Panel, Definitions). Both concepts can be
used in relation to individuals and groups. If you work
with a group of people, you can tell whether their morale
is high and whether they are motivated, but it can be
difficult to say exactly what it is you sense. Morale is to
do with confidence, trust, optimism, self-belief in oneself
and the others around you. It is a feeling about a general
situation. Motivation is a state of will. It is about a person’s
commitment to actually doing something to achieve a
particular goal.
A study of drivers of track
maintenance trains in Japan
found a close relationship
between measures of morale
and motivation on the one hand
and accident rates on the other: the
lower their morale and motivation, the
greater the risks of an accident.
Understanding Human Factors/June 08
Definitions
Motivation
‘Motivation derives from the Latin verb ‘movere’, meaning
‘to move’. Movement implies action and, in order to act,
energy and effort are required from the individual. The level
of individual motivation is determined by the amount of
energy and effort people put into their work.’
Kakabadse et al (1988)
Morale
‘Morale is a state of mind. It is that intangible force which
will move a whole group of men to give their last ounce to
achieve something without counting the cost to themselves;
that makes them feel they are part of something greater
than themselves.’
Field Marshal Slim, Defeat into Victory (1956)
Page 117
Part 2: Guidance
Conditions
Morale and motivation are not the same thing, then, but
social opportunities that work often provides. This could
they do tend to go hand-in-hand. It is hard to imagine
be important for a signaller working in a single-person
a person whose morale is low being highly motivated
signal box or a train driver driving trains on long routes.
to do a particular task – except, perhaps for reasons of
Or a person who is just doing a job on a temporary
safety and self-preservation. Most of the research and
basis, while they save up money to do something else,
theory in this area has focused on
may have little interest in the work
motivation, so we are mainly going
‘The beatings will continue itself, only being motivated by the
to deal with motivation here. You
financial rewards.
until
morale
improves.
’
can assume that the factors at work
that influence motivation (including Commander of the Japanese
How to motivate staff has long
Submarine
Fleet,
WW
II
your own decisions and actions) will
been a central concern for
probably also influence morale as
managers, and many theorists have
well.
put forward ideas about how to motivate people. We
begin with a class of theories about motivation that starts
from the common-sense assumption that people work to
What motivates people at work?
get their needs met.
Mullins (1999) says that people seem to work for three
main reasons:
1 Economic rewards – most people need to be paid, to
have a pension, to feel secure, and so on
2 Intrinsic satisfaction – most people want to do a
job that they find interesting and gives them the
opportunity to learn, to develop new skills, and to
progress
3 Social relationships – work is one of the main places
where many people make friends, feel part of a group,
have status, earn respect, exercise power, etc
Source: Mullins (1999), reproduced with permission
Most people are motivated to work because of some
mix of all three reasons, although the precise make-up of
the mix will vary. For example, somebody with a strong
family and social life outside work may not need the
Page 118
Theories about needs
One well-known theory of this type
is Maslow’s hierarchy of needs, shown
in the diagram, Maslow’s hierarchy of
human needs. Maslow argued that
people will only experience a particular
level of needs if the levels below it
are satisfied. For example, if you are
feeling hungry, tired or unsafe, you
probably will not be much concerned
about your needs for friendship and
belonging, and even less so about
your needs for personal growth. But in
practice, people do not always seem
to experience needs in this order and
will often sacrifice lower level needs to
achieve higher ones. For example, you
may be willing to miss meals or sleep in pursuit of some
social goal. But as a general classification of the kinds of
needs that can motivate people, Maslow’s list has proved
useful.
Another useful view of motivation is provided by
Herzberg. He identified a number of factors that
motivated people at work. He called these ‘motivators’
or ‘growth factors’. But it is not simply the absence of
motivators that causes dissatisfaction. Dissatisfaction is
caused by a separate group of factors he called ‘hygiene
or maintenance factors’. These two classes of factors
are shown in the Panel on page 119, Herzberg’s work
motivation factors.
There are numerous other theories of needs. (You can
find out about them in some of the sources listed in
Further information at the end of this section.) But the
Based on original research by Maslow (1943)
Understanding Human Factors/June 08
Part 2: Guidance
Conditions
Herzberg’s work motivation factors
Frederick Herzberg originally published his influential
two-factor theory of motivation in 1959. He
distinguished two classes of factor, one of which he
called motivation factors and the other hygiene factors.
Motivation factors include receiving recognition, being
given responsibility, having opportunities for personal
growth, and the like. The presence of these factors is
motivating, but if they are absent, then the result is not
usually a lack of motivation. Hygiene factors include
level of pay, job security, good working conditions, and
so on. The effect of these factors is the opposite to
that of the motivation factors. If hygiene factors are
lacking (poor pay, bad conditions, etc), then staff are demotivated. But if they are present, this does not result
in high motivation.
Summarised from: Herzberg (1959)
general conclusion from all the theories is that what
motivates people is a complex set of factors. Individuals
will differ from each other in what motivates them. The
same person may be motivated by different things at
different times. To motivate people means keeping in
close contact with them, constantly asking them questions
and listening to what they tell you.
Theories about expectations
A number of writers on motivation have focused less
on needs and more on the relationships between how
much effort people are prepared to put into their work,
the expectations that people have about being rewarded
for their efforts, and to what extent they see the reward
system as fair. People at work tend to observe carefully
what happens to them and to others around them. They
Understanding Human Factors/June 08
observe how much effort or what level of performance
leads to what kinds of reward. If they see that extra effort
or high levels of performance go unrewarded, this will
generally reduce motivation. Also if they perceive any
injustices in how rewards are allocated, this too will be
de-motivating.
How can you improve morale and motivation?
It may be that the system of reward is truly unfair or it
may be that staff are unclear about what is expected of
them in terms of satisfactory performance. It may also
be that the person does not have a realistic perception
of their own level of performance. Managers and
supervisors must be careful in making clear what they
expect from staff, and must also give them objective
feedback about how well they are doing.
• Your staff. Spend time talking with your staff, especially
listening to them. Find out what interests them and
what they want from their jobs.
An important idea here is the psychological contract. In
most employment situations there is an actual contract
between the employer and the employee. This sets out in
black and white what the rights and responsibilities each
has – pay, duties, periods of notice, and so on. But there
will also be a contract that is not written down but is
equally real. This psychological contract includes the many
and various expectations that the employee has of the
employer and vice versa. Such expectations might include
the opportunities for promotion, a generally pleasant
working atmosphere, minimum standards of politeness, a
freedom from unreasonable demands, and so forth.
Managers must be aware of the psychological contract
whenever they introduce any changes. Violations of
this contract will be seen as a breach of trust, and staff
motivation will be damaged. Details on where to find out
about these theories are given in Further information.
Motivation is ultimately about the question: why do
people do what they do? You will find as many answers to
this question as there are people you work with. Drawing
on the various motivation theories, the following are
guidelines on how to motivate staff.
• The jobs you are responsible for. Make sure you know
the rewards that tend to be associated with the
jobs of your staff. Some jobs, for example, offer few
opportunities for social contact. Or a job may in truth
be boring. For some people such things may not
matter much, but for others they will be highly offputting.
• Working conditions. Make the physical and
organisational conditions as attractive as possible.
At the very least ensure that they do not become
sources of dissatisfaction.
• Performance. Make sure everyone knows what is
expected of them. Give them plenty of fair and
objective feedback so that they know how they are
doing.
• Job enrichment. For those who indicate they want extra
variety or responsibilities, find ways to make their job
more varied. But do not assume everyone wants this;
some people might think you are just giving them
extra work.
Page 119
Part 2: Guidance
Conditions
• Development and career opportunities. Help those staff
who want to move ahead to find opportunities to
progress their careers.
• Psychological contract. Understand the implicit
expectations your staff have of the organisation, and
what the organisation expects of them. Many of
these expectations may not be directly related to
the work. For example, staff may expect working for
a certain organisation to give them a status in their
local community. Assess any changes you plan to
introduce for their impact on these usually unspoken
assumptions.
• Rewards. Ensure rewards are fairly allocated and
that staff fully understand the basis on which
rewards are given. Remember rewards are not
only financial. Recognition, praise, a good working
location, favourable working hours, even being given
an interesting extra job, and much else can be highly
motivating.
• The big picture. Keep staff up-to-date with how the
whole department, region, organisation, industry is
doing. Especially report areas of progress. Being able to
place their work in a bigger context helps individuals
feel their jobs are worthwhile and meaningful.
• Yourself. Look at the messages your attitudes, behaviour
and general demeanour send to your staff. If you
are confident, optimistic and enthusiastic, this will be
infectious and motivating to those around you.
Page 120
Further information on morale and motivation
Stress
1 Adair J. (2004) Adair on Teambuilding and Motivation,
Thorogood
What is stress?
2 Fournies F. (1999) Coaching for Improved Work
Performance, McGraw-Hill Education
3 Handy C. (1993) Understanding Organisations, 4th
Edition, Penguin
4 Harvard Business Review (2003) on Motivating
People, Harvard Business School Press
5 Heil G. & McGregor D. (2000) Revisited - Managing
the Human Side of the Enterprise, John Wiley
6 Herzberg F. Mausner B. & Snyderman B.B. (1959) The
Motivation to Work (2nd ed.) John Wiley, NY
7 Herzberg F. (1993) Motivation to Work, Transaction
Publishers
8 Kakabadse A. et al (1988) Working in Organizations,
(especially Chapter 5), Penguin
9 Maslow A. (1943) A Theory of Human Motivation,
Psychological Review, 50, 370-396
10 Maslow A. (1998) Maslow on Management, John Wiley
There are many definitions of stress. The definition and
description used by the HSE is:
‘Stress is the adverse reaction people have to excessive
pressure. It isn’t a disease. But if stress is intense and goes
on for some time, it can lead to mental and physical ill
health.’
The HSE says that stress is widespread in the British
workplace:
• About 1 in 5 people say that they find their work
either very, or extremely, stressful.
• Over half a million people report experiencing workrelated stress at a level they believe has actually made
them ill.
• Each case of stress-related ill health leads to an
average of 29 working days lost.
• A total of 13.4 million working days were lost to stress,
depression and anxiety in 2001.
• Stress has overtaken back problems as the leading
complaint in disability claims.
In a safety critical industry such as the railways, stress is
a particular issue because stress makes people less safe
in their work. Many people being absent from work
because of stress only adds to the problem by increasing
the workload (page 125) and associated stress of those
people still at work.
Understanding Human Factors/June 08
Part 2: Guidance
Conditions
What are the effects of stress?
Stress happens when the safe limits of our natural arousal
levels are exceeded. In fact, there is a well-established
relationship between performance at work and stressful
demand. When the demands placed on us are very low,
we tend to perform badly out of boredom or frustration.
As the demand increases, our level of arousal rises and
we become more alert and decisive. At some medium
level of arousal – which varies from person to person –
we perform at our best. We feel in total command of our
work, able to cope with all the challenges it can throw
at us. However, if the demands continue to increase and
are not balanced by periods of lower demand, when we
can recover and rebuild our reserves of energy, we begin
to be stressed. We start to rush tasks, we don’t check
properly, we don’t consider alternatives and we make
more and more mistakes. (Stress is often cited as a cause
of SPADs.) If the level of stress continues to increase,
we may eventually reach a stage called burnout (see
supervision and appraisal on page 68 for how to recognise
and deal with this), where we are ill and unable to work.
It is important to note that while people need to be
aroused to some extent in order to act, stressors are
always bad (see Panel on page 122, What causes stress
at work). For example, even a little anxiety (eg caused by
fear of bullying at work) affects performance by producing
changes in people’s ability to pay attention, learn and
recall information.
Understanding Human Factors/June 08
Stress is usually first experienced psychologically, typically
with difficulties in sleeping, increasing anxiety and feelings
of tension and loss of control. Problems can then start to
appear at work, with difficulties in concentrating, inability
to make decisions, becoming increasingly short-tempered
with colleagues.
disappears with each person who leaves. More resources
have to be devoted to recruitment (page 84) and training
(page 55) to replace the staff who leave. But replacing
staff who leave because of stress is only a temporary fix.
Something needs to be done about the underlying causes
of stress.
But when levels
of stress are
‘Too much stress
maintained for
and we burn out …
any length of
time, physical
but there are many
symptoms will
begin to emerge. symptoms that can let
Headaches and
us know it’s happening.’
backaches are
common signs
of stress for many people. Stress tends to seek out a
person’s particular areas of weakness. For one person,
this may mean digestive problems. For another person
stress may show up in skin rashes.
What causes stress?
While one should avoid the trap of blaming any physical
ailment on stress, whenever these kinds of problem
appear consideration should always be given to the
possible role of work-related stress.
Clearly, stress has a bad effect on the individual. In
addition, stress is also a problem for the workgroup as
conflicts increase, and cooperation and relationships
suffer. All of these effects reduce the quality and safety of
the work being carried out. At the organisational level,
stress results in a fall in performance and productivity.
Staff turnover increases, and experience and expertise
Stress has many possible causes (see Panel on page 122,
What causes stress at work?). A recent survey by the TUC
identified workload (page 125) as far and away the major
cause of stress. Other leading causes were cuts in staff,
change, long hours, bullying and shift work (page 128). On
the railways the factors that cause stress include pressure
to achieve targets, threats from aggressive or violent
passengers, and risks posed by vandals or trespassers.
Suicides are a particularly severe cause of stress for a
significant minority of train drivers. RSSB has recently
conducted research aimed at minimising the impact of
railway suicides on railway staff (see Further information).
The HSE has carried out extensive work in the area of
stress at work in recent years. One key output of this
work is the set of Management Standards relating to
stress (see Further information at the end of this section).
The standards classify the principle causes of work-related
stress into six key areas:
• Demands – includes such issues as workload, work
patterns and the work environment
• Control – how much say the person has in the way
they do their work
• Support – includes the encouragement, sponsorship
Page 121
Part 2: Guidance
Conditions
What causes stress at work?
The causes of stress are very personal. We all vary in
what we find stressful. What is enlivening music for one
person may be unbearable noise for another. Our age,
gender, experience, home life, and many other factors will
influence our vulnerability to different types of stressor.
Somebody who is having a difficult time in their home life
is likely to be more susceptible to stressors at work. But
we can identify certain classes of stressor that are likely to
be behind most cases of stress.
First and most obviously, physical conditions can give
rise to stress. Loud noise, poor lighting, vibration from
passing trains or road traffic, too much or too little heat
can all contribute to stress. Poorly designed workspaces
which are cramped or where equipment and furniture is
placed so that it is easy to bump into or trip over, can be
psychologically stressful as well as dangerous. Equipment
that is difficult to access, or difficult and frustrating to use
are common contributors to stress levels of staff.
Perhaps most commonly, stress is associated with levels of
work, especially pressure of too much work and too little
time, although too little to do can also contribute to stress.
Uncertainty is a major element in many cases of stress at
work. This may be uncertainty about what to do or what
is expected of you, how well you are doing, what your
promotion prospects are, and so on. Lack of job security is
a serious cause of stress.
For many people stress is associated with their
relationships with colleagues, managers and even
subordinates. Bullying, conflicts and harassment are all
acute causes of stress.
Page 122
and resources provided by the
organisation, line management and
colleagues
• Relationships – includes promoting
positive working to avoid conflict
and dealing with unacceptable
behaviour
‘Stress at work is a
major cause of poor
performance … and the
major cause of stress at
work is workload.’
• Role – whether people understand their role within
the organisation and whether the organisation ensures
that the person does not have conflicting roles
• Change – how organisational change (large or small) is
managed and communicated in the organisation.
Note that not all these events are
‘negative’. Some of them are events
that are usually considered as
‘positive’ (eg getting married). Even
the ‘good things’ in life can impose
demands that may be experienced
as stressful.
Stress within work and stress outside of work tend to
add together. This means that moderate levels of stress at
work and at home may mean that the overall experience
of stress is high. But equally a happy, settled home life may
mean that the impact of high levels of stress at work is
minimised.
Source: HSE (2005), reproduced with permission
Stress is also, of course, caused by many life events
outside of work. The events most strongly associated with
high levels of stress are:
How can you recognise stress?
• death of partner, spouse or other close relative
Good practice in stress management begins with
recognising stress in yourself or in others around you
at work. Things to look for particularly in your own
behaviour include:
• divorce
• eating on the run, or in a disorganised manner
• marital separation
• smoking, or drinking excessively
• imprisonment
• rushing, hurrying, being available to everyone
• personal injury or illness
• doing several jobs at once
• getting married
• missing breaks, taking work home with you
• dismissal from work
• having no time for exercise and relaxation.
• marital reconciliation.
Source: HSE (2005), reproduced with permission
Source: Holmes & Rahe (1967), reproduced with permission from Elsevier
Understanding Human Factors/June 08
Part 2: Guidance
Conditions
Aspects of behaviour that you may notice about your
colleagues include:
• changes in a person’s mood or behaviour
• deteriorating relationships with colleagues
• irritability
• indecisiveness
your employer has a legal responsibility for your health at
work and this includes stress arising from work. Maybe
some of your stress arises from fears that you do not
have all the knowledge and skills for your work. Perhaps
there are training courses you could attend to help you
develop new knowledge and skills. On a wider scale you
may need to re-appraise all of the commitments you have
within work and outside work. Are you just taking on
more than is sensible?
has a stress problem before they are fully aware of it
themselves. The Panel on page 124, Looking after others,
gives good practice for stress prevention and initial stress
management if stress does become a problem. The
suggestions given in the section on Helping yourself (this
page) may also give you ideas that you can pass on to
others who ask for your assistance.
However effective you become at managing your
external demands, the chances are that there will still be
times when demands will be uncomfortably high. This is
why you need to improve your own inner resources as
well. When you look at your own ability to cope with
stress, you may realise that you need to develop skills that
are not directly work-related and are of a more personal
nature. Time management is one area you might want
to improve. Or learning how to say ‘no’ when asked to
take on just one more responsibility. It may be that you
want better control over how your body reacts to stress.
Learning how to relax, to take more exercise, to eat
more healthily, to take more time off, may all be things
you want to consider. You might also want to find out if
your company offers training in stress management or
counselling. The sources listed in Further information at the
end of this section will help you find out what you can do
to improve your ‘stress proofing’.
‘Stress is often a symptom of poor employment relations
and can seriously affect productivity. Organisations who
talk regularly with their employees and have sound
systems and procedures in place for dealing with issues
like absence and discipline are much more likely to avoid
work-related stress and to be able to deal with potentially
stressful situations when they arise.’ ACAS
• absenteeism
• reduced performance
• smoking or drinking alcohol more than usual
• indications of drug abuse
• complaints about their health.
Source: HSE (2004), reproduced with permission
See supervision and appraisal (page 68) for more on how
to recognise and manage stressed colleagues.
How can you reduce stress?
Helping yourself
Stress arises when you experience a mismatch between
the demands placed on you, and your capacity (in terms
of time, energy, skills and so on) to cope with them. To
manage your stress you need better control over the
demands you face and improved capacity to cope.
If you are responsible for the work and well-being
of others in your organisation, you need to do more
than respond to particular causes of stress and adopt
a more formal and comprehensive approach to stress
management. HSE requires every employer to conduct
risk assessments for health and safety hazards, including
work-related stress. It recommends a five-step approach
to risk assessment.
• Identify the hazards
• Decide who might be harmed and how
Helping others
Gaining control over the demands you face at work may
mean re-assessing what you can and cannot do. This in
turn may lead you into re-negotiating with managers and
colleagues what responsibilities you take on. Remember
Helping your organisation
If you are responsible for the work or well-being of
others in your organisation, you will often be in the
best position to recognise and help someone suffering
from stress. You may well be aware that the person
• Evaluate the risk and take action
• Record your findings
• Monitor and review
Source: HSE (2005), reproduced with permission
Understanding Human Factors/June 08
Page 123
Part 2: Guidance
Conditions
Looking after others
What can you do to prevent stress from becoming a problem?
•Show that you take stress seriously, and be understanding towards people who admit to
being under too much pressure.
•If you are a manager, have an open and understanding attitude to what people say to you
about the pressures of their work, and look for signs of stress in your staff.
•Ensure that staff have the skills, training and resources they need, so that they know what
to do, are confident that they can do it and receive credit for it.
•If possible, provide some scope for varying working conditions and flexibility, and for
people to influence the way their jobs are done. This will increase their interest and sense
of ownership.
•Ensure that people are treated fairly and consistently and that bullying and harassment
aren’t tolerated.
•Ensure good two-way communication, especially at times of change.
•Don’t be afraid to listen.
What should you do if an employee complains about being stressed?
•First, listen to them! If the stress is work-related:
• try to address the source(s)
• involve the employee in decisions
• if necessary, encourage them to seek further help through their doctor
• if not their manager, ensure that they are treated with understanding and in confidence.
•Where you can’t control the work-related sources of stress, it may be appropriate to
move the employee if you can. If a period of sick leave is recommended, keep in touch
with the employee and their doctor. Remember that before they are ready to return to
their old job, they may be able to return to work to do part of their job, work reduced
hours or do a different job.
•Try to be flexible!
•Don’t be tempted to think that firing someone provides an easy way out! If you don’t act
reasonably in dismissing an employee, they could claim unfair dismissal.
•Finally, bear in mind that if one of your employees is suffering from work-related stress,
they may represent the tip of an iceberg. Find out whether others are also experiencing
stress at work.
This approach is also based on
the new Management Standards
for work-related stress that
we mentioned earlier. The HSE
website is the best place to get
full information about both the
standards and the assessment
process.
Further information on stress
1The HSE provides many
resources dealing with stress
at work. The web page URL is
www.hse.gov.uk/stress/index.htm
(as of May 2008)
2www.hse.gov.uk/stress/
standards/index.htm HSE Stress
Standards (as of May 2008)
6 The TUC has a useful account of stress and a
description of the management standards from a
union perspective: www.tuc.org.uk/h_and_s/tuc10147-f0.cfm (as of May 2008). Most unions involved
in the railways also provide guidance on stress, usually
referring to the HSE and TUC websites
There are also many self-help books on stress, eg
7 Davis M. et al (2003) The Relaxation and Stress
Reduction Workbook, New Harbinger Publications
8 Greener M. (2003) The Which? Guide to Managing
Stress, Which? Books
9 RSSB (2005) Minimising the Impact of Railway Suicides
on Railway Staff, Research Project T317.
10 Richards M. (1998) The Stress Pocketbook,
Management Pocketbooks
3HSE (2004) Work-related
stress: A short guide
11 Sutherland V. & Cooper C. (2000) Strategic Stress
Management: An Organizational Approach, Palgrave
MacMillan
4HSE (2005) Tackling stress: The
management standards approach
12 Williams A. & Cooper L. (2002) Managing Workplace
Stress: A Best Practice Blueprint, John Wiley
5International Stress
Management Association, ACAS,
HSE, Working together to
reduce stress at work: A guide
for employees, 2004
Two other useful books are:
13 Covey S.R. (1999) The Seven Habits of Highly Effective
People, Simon & Schuster
14 Pedler M. Burgoyne J. & Boydell T. (2001) A Manager’s
Guide to Self-development, McGraw-Hill Education
Adapted from HSE (2004), reproduced with permission
Page 124
Understanding Human Factors/June 08
Part 2: Guidance
Conditions
Workload
What is workload?
How much workload is too much?
Physical workload
Workload refers to the effort demanded from people by
the tasks they have to do. It can be the effort demanded
at a single point in time, or over a whole shift. It can be
the physical demands created by working in a particular
posture (eg sitting, standing or reaching at a workstation);
manual labour (such as walking, using hand tools, carrying
loads); or working in particular environmental conditions
(eg extremes of temperature and humidity, and poor
lighting). Workload can also be the mental demands
created by the need to attend to sources of information
and then process the information, often against time
pressure. Mental workload can be considerably increased
by the operator’s knowledge that they are responsible
for processing information correctly – and what the
consequences of error might be.
Excessive physical workload arises from the following
typical problems:
Workload is a crucial human factors consideration for
both designers and managers. If people have too little to
do, they can become bored and inattentive. On the other
hand, too much workload is a primary cause of high
stress, which seriously degrades human performance. In
either case, there is a major implication for safety-critical
operations.
• a task is performed under time pressure
‘Workload is a problem for
safety-critical operations if it’s
too low – and an even bigger
one if it’s too high.’
Understanding Human Factors/June 08
• a load is too heavy and/or bulky, placing unreasonable
demands on the person
• a load has to be lifted from the floor and/or above the
shoulders
• a task involves frequent repetitive lifting
• a task requires awkward postures, such as bending or
twisting
• a load cannot be gripped properly
• a task is performed on uneven, wet, or sloping floor
surfaces
• a task incorporates too few rest breaks.
These problems may result in physical injuries (eg to the
lower back, arms, hands or fingers) and increase the risk
of slips, trips and falls.
Much is known about the measurement and limits of
physical workload. You can find specific guidance in the
Further information at the end of this section.
Mental workload
Excessive mental workload arises most typically in
tasks that demand more attention than operators
have available. People continually change their working
methods to keep their mental workload within limits.
They may, for example, vary the level of detail they attend
to in an attempt to keep the whole picture in their heads.
Problems arise, however, when they pitch their attention
at too high a level and so completely miss important
events or trends in the workflow. Alternatively, sometimes
people get tunnel vision as they focus on one particular
task element in detail – and so fail to address the rest of
their responsibilities. It is possible to see this happening.
For example, you may see an air traffic controller who is
becoming overloaded in this way move their face closer
and closer to the display screen as they slowly ‘lose the
picture’.
When mental workload gets too much, people become
stressed (see Stress, page 120) and their accuracy suffers.
They give more attention to tasks or information that
they consider to be important (which may or may not
be the case). They focus on information sources that are
easiest to see. They use strategies that require the least
mental effort – these are usually the best learned, but
may not be the most appropriate. They may also get
locked into a single strategy.
Low workload can also be a major problem. Over time
it can reduce a person’s ability to notice new events and
result in slower response times. Sometimes people just
go to sleep on the job. People are more likely to become
sleepy through boredom if they have suffered sleep loss
or disruption to sleep rhythms through shift work (page
128).
Page 125
Part 2: Guidance
Conditions
People vary a great deal in their
capacity for, and vulnerability to, high
workload. Organisations can minimise
the negative effects of workload
through the following strategies:
• Design (page 25) (Interfaces,
Workplaces, and Jobs) – to
minimise the occurrence of work
over/underload.
‘Mental workload
problems arise most
often when tasks exceed
the amount of attention
that operators have
available.’
• Training (page 55) – including the use of ‘overtraining’;
and creating an appropriate team culture that means
people anticipate workloads and arrange appropriate
backup – see teamworking on page 103.
Note: ‘Overtraining’ simply refers to giving people more
training than they apparently need to pass the test at the
end of training. It works because it gives people a chance
to practise what they have learned so that their skills
become more automatic. Such a strategy is especially
useful when training people to respond very rapidly to
emergencies.
is by no means ideal as well as being
expensive. It is important to use
workload prediction techniques at
an early stage of design to identify
likely problem areas, determine the
use of automation and define user
roles. Workload measurement should
be used during prototyping to help
identify design deficiencies and/or the
need for user role re-definition.
A natural approach to reducing workload at the design
stage is to find ways to automate some tasks. However,
you need to take great care here, since this approach
often transfers the problem elsewhere (see function
allocation on page 37).
How do you measure workload?
Physical workload
Kroemer & Grandjean (1997) say the best indicator of
physical workload is heart rate, which rises with increasing
workload. This rate of rise is steeper:
• Selection (page 79) – to assure the correct mix
of operator ability and design solution to prevent
operators becoming over/underloaded.
• the higher the ambient temperature
• Recruitment (page 84) – to assure the supply of the
right numbers of human resources to perform the
required tasks.
• the smaller the number of muscles involved.
Design is the most important of these strategies. But if
a user-centred design (page 25) strategy has not been
used, you will usually have to fall back on one or more
of the other human factors areas to find a solution. This
Page 126
• the greater the proportion of static to dynamic effort
Kroemer & Grandjean (1997) reproduced with permission
The least intrusive way of measuring
this in a person is for them to wear
a continuously recording heart
monitor.
A toolset for calculating safe limits for manual handling
operations is the HSE Manual Handling Assessment Chart
(MAC) (see Further information).
Mental workload
There are a number of workload prediction and
measurement techniques – ranging from rating scales to
performance monitoring. You need to select the most
appropriate of these for use in the design process.
Primary inputs for workload analysis are:
• task analysis (page 47) • design scenario analysis
• role definition • team design (page 103).
These inputs together define the nature of the workload
(frequency of scenario events and user tasks to be
performed with the equipment) and the resources
available (user roles, numbers and team organisation).
You can measure or estimate workload using:
• timelines – which can also be used as the basis for
measuring real-time workload during prototyping
• human performance models via task simulation
software such as SAMMIECAD or IPME
• performance measurement (eg throughput time,
number of work units processed, error rate, task
conflicts observed etc.)
‘Design is the single best
way to avoid problems
of high workload – if it’s
not too late.’
• subjective rating scales.
Understanding Human Factors/June 08
Part 2: Guidance
Conditions
You will find summaries of some of the more useful
mental workload analysis tools in Part 3 of this Guide.
They include:
• ISA – Instantaneous SelfAssessment of workload
technique, a very simple
subjective technique that
was developed for use in
the assessment of mental
workload during the design
of future systems
You can use the results of workload analysis to identify
achievable levels of workload and as the basis for many
decisions about function allocation
‘It is easiest to estimate
(page 37), task and role design,
team design and user-equipment
mental workload via selfinterface design. The results will
report – and there are some also be used in workspace and
simple and effective tools to workstation design and the health
and safety analysis. See the section
do so.’
on Design (page 25) where many
of these issues are considered.
• SWAT – Subjective Workload Assessment Technique,
one of the most widely used and well known
assessment techniques available
• Team Workload Assessment – in which team members
provide a subjective assessment of their own workload,
as well as an estimation of the team’s overall workload.
• IPME – The Integrated Performance Modelling
Environment – software in which a task simulation
network is set up. When the software is run, the
simulated behaviour is influenced by software models
of factors that affect human performance (such as
noise and vibration). The results show how task timings
and accuracy alter with different levels of workload.
In addition, RSSB has recently developed a suite of tools
to measure train driver mental workload (see Further
information) and Network Rail is developing similar tools
to measure signaller workload.
Understanding Human Factors/June 08
• assessing employees’ work by considering the following
questions:
Do they work in a comfortable position?
Do they complain of any discomfort, including (aches,
pain, fatigue, or stress?
Are they satisfied with their working arrangements?
Is the equipment appropriate, easy to use and well
maintained?
Are there frequent errors?
How do you identify workload problems?
The most powerful way of stopping workload problems
from developing in the first place is to ensure there is
good design. This may entail using workload assessment
techniques and/or structured interviews with users on
prototype interfaces, workstations, work schedules and
workflows. User trials are particularly important. Mental
demands, in particular, are often overlooked during the
workplace design (page 41) process because they are
invisible and not an obvious part of the equipment design
(page 31).
After the design stage, the best
methods for identifying workload
problems are:
• talking to employees and getting
their views – they are very well
informed about the problems they
have!
Are there signs of poor or inadequate equipment design,
(page 31) such as plasters on employees’ fingers or
‘home-made’ protective pads made of tissue or foam?
Do the accident report book, the absence record or staff
turnover give any clues about any systematic problems?
• deciding if any on-job problems require deeper
investigation via the use of a workload assessment
technique.
During the process of identifying workload problems
you will often immediately see ways of eliminating them.
A minor alteration in the workplace
‘Sometimes a minor
may be all that is needed. But you
will need to make sure that any
alteration in the
alterations are properly evaluated by
workplace may be all
the people who do the job. Be careful
that’s needed to reduce that a change introduced to solve one
problem does not create difficulties
workload to safe levels.’ elsewhere.
Page 127
Part 2: Guidance
Conditions
Further information on workload
Physical workload
to the human: A textbook of occupational ergonomics
(Fifth edition) Taylor & Francis
1 HSE (2000) Management of Health & Safety at Work
Regulations 1999 Approved Code of Practice and
Guidance (L21) (Second Edition) HSE Books
12 HSE (1998) Manual handling. Manual Handling
Operations Regulations 1992. Guidance on
Regulations L23 (Second edition) HSE Books
2 HSE Musculo-Skeletal Disorders
www.hse.gov.uk/msd/faq.htm (as of May 2008)
13 Pheasant S. (1991) Ergonomics, work and health,
Macmillan
3 HSE Manual Handling Assessment Charts (MAC),
www.hse.gov.uk/pubns/indg383.pdf (as of May 2008)
Mental workload
4 HSE (1994) Manual handling: Solutions you can
handle. HSG115
5 HSE (2000) Getting to grips with manual handling: A
short guide for employersLeaflet INDG143(rev1)
6 HSE (2003) Aching arms (or RSI) in small businesses
- Is ill health and sickness absence due to upper
limb disorders a problem in your workplace? Leaflet
INDG171(rev1)
7 HSE (1997) Seating at work HSG57 (Second edition)
8 HSE (1997) Lighting at work. HSG38 (Second edition)
9 HSE (2003) Work with display screen equipment.
Health and Safety (Display Screen Equipment)
Regulations 1992. Guidance on Regulations (Second
edition) L26
10 HSE (1998) Working with VDUs Leaflet INDG36(r1)
www.hse.gov.uk/pubns/indg36.pdf (as of May 2008)
11 Kroemer K.H.E. & Grandjean E. (1997) Fitting the task
Page 128
1 ISO 10075-1:2000. Ergonomic principles related to
mental workload: General terms and definitions
2 RSSB Mental Workload Assessment in the Rail
Industry, Research Project T147
www.rssb.co.uk/pdf/reports/research/T147%20
Train%20driver%20mental%20workload%20-%20
the%20train%20driver%20workload%20principles%20
guidance%20note.pdf (as of May 2008)
3 Wilson J.R. & Corlett E.N. (1995) Evaluation of human
work: A practical ergonomics methodology (Second
edition) Taylor & Francis
Shift work
Shift work is inevitable in many parts of the railway
industry. A lot of attention has been paid in recent years
to its effect on train drivers. But, of course, many other
railway employees have to work shifts and possibly suffer
from the effects of doing so. Since 2003, the Hidden
Recommendations have set maximum limits for railway
shift working following the Clapham rail crash. These limits
are based on what appeared to be common sense at the
time, but are not based on scientific research. Since then,
further direction has been given by the EU Working Time
Directive. Poorly managed shift patterns can have wideranging effects on the staff concerned. They can also affect
the efficiency and, most critically, the safety of the industry
as a whole. It is therefore important to understand the
risks and problems associated with different shift work
patterns.
What’s the problem with shift work?
Working shifts can have many side-effects. The main ones
are listed here.
• Fatigue. One of the most inevitable and damaging
consequences of shift work is fatigue. This has been
explicitly recognised on the railways since the time of
the Hidden inquiry into the 1988 Clapham Junction
accident. Fatigue is the subject of Regulation 25 in the
Railways and Other Guided Transport Systems (Safety)
Regulations 2006 (known as ROGS) which came into
force on 10 April 2006:
“Every controller of safety critical work shall have in
place arrangements to ensure, so far as is reasonably
practicable, that a safety critical worker under his
Understanding Human Factors/June 08
Part 2: Guidance
Conditions
rushed or interrupted. Shift workers tend to rely more
management, supervision or control does not carry out
on snack foods with a high fat content, and to drink
safety critical work in circumstances where he is so
more coffee to stay alert.
fatigued or where he would be liable
to become so fatigued that his health
‘Shift work creates
Cardiovascular problems tend also
or safety or the health or safety of
to be more prevalent amongst
other persons on a transport system several problems for
shift workers. Again, the type of
could be significantly affected.
people, but the biggest meals consumed and the lack of
is likely to contribute to this.
The arrangements in paragraph (1)
by far is fatigue – leading exercise
Furthermore,
it has been reported
shall be reviewed by the controller
to increased risk of
that some women workers can
of safety critical work where he has
adverse effects on
reason to doubt the effectiveness of
mistakes and accidents’ experience
hormonal
and
reproductive functions.
those arrangements.”
The HSE says fatigue from shift work and overtime
is in the Top Ten topics for onshore HID (Hazardous
Installations Directorate) industries. Fatigue levels
clearly need to be monitored very carefully in shift
workers, particularly those playing a safety critical role.
It is generally recognised that on average adults need
7–8 hours sleep per night. Less sleep than required
will incur a ‘sleep debt’. A sleep debt can lead to
impaired alertness, which will adversely affect fatigue
and reaction times, concentration and judgement and
decision making. Sleep debt is accumulative and over
several days the effects can be compounded.
• Health problems. Gastro-intestinal problems can be
common among shift workers. Peptic ulcers and other
stomach disorders are five times higher among shift
workers with night-shifts, compared to day workers
or shift workers without night-shifts. The most likely
reasons for this are that shift workers tend to take
meals at irregular times and these meals are often
Understanding Human Factors/June 08
• Stress. Shift work can undoubtedly increase workrelated stress. This will particularly be the case when
the employee has little control over the shifts worked,
when shifts are unpredictable, where hours are long
and unsociable, and where inadequate provision is
made for rests and breaks.
• Family problems. Family life is often centred around
the shared rhythms of sleep, mealtimes, work and
recreation. Shift work can bring a separation from
these shared rhythms of general life and can isolate the
shift worker from family and friends.
• Anxiety, irritability and depression. These are more
commonly reported amongst shift workers. In many
cases they will clearly be inevitable consequences
of the side-effects of shift work already discussed.
However, they will also occur when a person feels they
have no choice but to accept shift work.
How can you reduce shift work problems?
The main problem with shift work is that of fatigue. In
March 2005 a major research study of shift work and
fatigue in train drivers was completed for RSSB. The study
identified a number of primary influences of shift work
on fatigue and recommended a series of guidelines for
reducing these influences.
Other key studies have also identified strategies for
dealing with fatigue induced by shift work, as well as a
number of its other side effects.
In the Panel on page 130, How to manage shift work
problems – a research perspective, we have used these
various sources to summarise the key shift work
problems and strategies for their management. These
guidelines are highly consistent with the advice published
by ASLEF (see Further information).
There are a number of tools and techniques for assessing
the fatigue risks associated with shift working patterns.
The guidance for managing fatigue in safety critical work
with reference to Regulation 25 of the ROGS (available
from the ORR) is a key document for anyone on the
railways who manages, supervises or controls workers
doing safety critical work. It is required that effective
arrangements are established for managing the risks of
fatigue in safety critical workers. This management process
should include the nine steps shown in the Panel on page
131, Managing the risks arising from fatigue in safety critical
workers. You can find detailed descriptions of what to do
at each of the steps in the ROGS Guidance. In working
through this process you will probably find it helpful to
make use of the Fatigue Risk Index tool from the HSE.
Page 129
Part 2: Guidance
Conditions
How to manage shift work problems – a research perspective
This tool uses the five main
factors that affect fatigue:
Shift work problem
Management
Long shift duration is a primary cause of fatigue. Its effect is
most significant on night and early morning shifts.
•Operate a maximum shift length of 8–10 hrs for nights and early shifts.
1 Shift start time
Long periods of continuous duty without a break are
particularly fatiguing. This can be especially so for drivers.
•Ensure breaks are taken of at least 15 minute duration, free from any work-related activities and that continuous driving
time is restricted to 4 hours to reduce accident risk.
2 Shift duration
High percentage of hours worked per week is very fatiguing
due to long shifts and short breaks between shifts – or both.
•Ensure a maximum of 55-60 hours per week for train drivers. This is consistent with a 55-hour rolling week limit for
aircrew and a 56-hours per week limit for HGV drivers.
Consecutive shifts increase accident risk due to fatigue. The
risk is greatest on consecutive night and early shifts due to
cumulative sleep loss. Long consecutive shifts can result from
workers volunteering to work rest days and Sundays.
•Ensure a maximum of 7 consecutive shifts before a rest day, with this reduced to 3 consecutive shifts for nights.
•Review policy on napping. Napping during the duty period is permitted in a number of safety-critical industries. For
example, airlines permit aircrew to take a 40-minute nap followed by a 20-minute recovery period during the cruise
phase of the flight. Canadian and American railway companies have also introduced schemes allowing napping on duty.
Nappers must be made aware of the recovery period required to overcome grogginess. In general, grogginess tends
not to be a major problem with naps of less than 30 minutes.
Shift variability is often inevitable but can be fatiguing. The
fatigue risk is greatest when there is a rapid switch from a
late finish or night shift to an early shift. Shift variability is
often increased by shift swapping between staff.
•Eliminate the requirement for shift variability and the opportunity for fatigue-inducing shift swapping.
Lack of rest between shifts is an obvious and common cause
of shift work-related fatigue.
•Ensure a minimum rest period of 14 hours between night shifts and 12 hours between other shifts.
Lack of rest days is another common cause of shift workrelated fatigue.
•Ensure a minimum of 2 rest days at the change from night to early shifts, and one rest day at the change from late to
early shifts. Ensure one rest day after seven consecutive shifts.
Health problems associated with shift work can build up over
prolonged periods of time (eg due to poor diet, weight gain,
heart problems and disrupted sleep patterns)
•Ensure a continuing programme of education is in operation eg on the use of exercise and how to manage sleep at
home eg exercise is often beneficial before the start of a late or night shift.
Lack of social interaction in late and night shifts exacerbates
shift work-related fatigue.
•Social interaction is not always easy to address (eg one-man signal boxes), but it does provide stimulation and help
counter fatigue.
Lack of work variety exacerbates shift work-related fatigue
– especially on late and night duties.
•Providing variation in the work across a shift can help combat fatigue, but is not always easy. Where possible, make sure
that employees are deterred from leaving tedious or routine tasks to the end of a shift when they are likely to be most
drowsy.
Inadequate temperature, ventilation and lighting promote
fatigue – especially shift work-related fatigue.
•Adequate lighting, proper ventilation and a comfortable working temperature will help control fatigue. A well-lit
workplace signals to the body that it is time to be alert and awake.
Poor diet can arise from difficulties with eating at the right
times or the right foods.
•Ensure there are facilities that enable – and a culture which encourages – regular, nutritious meals to be taken.
Sources: RSSB research and general human factors literature
Page 130
•However, be aware that allowing employees to select/swap shifts to tailor work to their needs can help reduce fatigue
and can improve satisfaction with the shift work system.
•Ensure a continuing programme of health monitoring for the specific risks associated with shift work is in operation.
•Caffeine can be a useful stimulant before and early on in a shift but should be avoided late in a shift if the employee is
due to sleep shortly.
•Ensure an education programme that teaches and reinforces good dietary practice, eg meals taken at the end of a shift
should be easily digestible so that sleep is not disturbed due to an active digestive system.
3 Length of time between shifts
4 Breaks within duties
5 Number of consecutive shifts.
It provides scores for individual
shifts, and these are then added
up to give the total score for
the roster pattern as a whole.
It can thus be used to compare
different shift patterns.
In addition to its use on the
railways, The Health and Safety
Laboratory has used the HSE
Fatigue Index to compare fatigue
levels associated with an 8-hour
shift rota and a proposed 12-hour
rota in other safety critical
environments, including the
offshore, nuclear, transport and
chemical industries.
RSSB has published a series of
resources aimed at combating
fatigue through shift work – see
Further Information.
Understanding Human Factors/June 08
Part 2: Guidance
Conditions
Managing the risks arising from fatigue in safety critical workers (ROGS 2006)
Further information on shift work
1 ASLEF Shift work, lifestyle and
health, ASLEF booklet
2 The London Chamber of
Commerce and Industry: 24/7
Health effects: shift and night
working – an employer’s guide,
Fact Sheet 10 July 2004
3 McGuffog A. Spencer M.B. Stone
B.M. & Turner C. (2005) Guidelines
for the management and reduction
of fatigue in train drivers, for RSSB
4 RSSB (2005) ‘Feeling tired’
resources. As of May 2008 www.rssb.co.uk/pdf/Feeling%20
tired%20additional%20info.pdf
5 Railway Safety (1996), Railway
(Safety Critical Work) Regulations
1994. Approved Code of Practice,
HSE Books
8 Wedderburn A. (ed) (2000) Shift work and Health,
Bulletin of European Studies on Time (BEST 1/2000),
European Foundation for the Improvement of Living
and Working Conditions
www.eurofound.europa.eu/publications/htmlfiles/
ef0009.htm (as of May 2008)
9 The Railways and Other Guided Transport Systems
(Safety) Regulations 2006 can be accessed from the
Office of Public Sector Information (OPSI) website at:
www.opsi.gov.uk/si/si2006/uksi_20060599_en.pdf ( as
of May 2008)
10 The Railways and Other Guided Transport Systems
(Safety) Regulations 2006 – Guidance on Regulations
can be accessed from the Office of Rail Regulation
(ORR) website at: www.rail-reg.gov.uk/server/show/
ConWebDoc.7964 (as of May 2008)
11 The development of a fatigue/risk index for
shiftworkers, to be published by HSE Books in 2006
6 Strategic Rail Authority (2003)
Guidelines for the Implementation
of the Working Time Directive,
Aug 2003
7 Stone B.M. (2004) Evaluation of
current tools and techniques used
for estimating risks associated with
shift patterns, QinetiQ Centre for
Human Sciences Report, QinetiQ/
KI/CHS/CR032327, for RSSB
Understanding Human Factors/June 08
Page 131
understanding
© Four by Three 2003
Techniques
human factors
understanding
Techniques
human factors
© Four by Three 2003
Part 3: Reference
Techniques
Part 3: Reference
Techniques
Parts 1 and 2 of this Guide describe a wide range of
human factors issues that managers, designers, trainers
and others on the railways are likely to encounter. They
contain specific guidance and tools on how to deal with
these issues. In addition, there are a large number of
human factors techniques that can be used to answer
particular questions.
In this section, a number of these techniques are
introduced. They have been selected because you may
find them of particular value in dealing with human
factors issues on the railways. For each technique, you will
find a description of when each technique might be of
interest and what is involved in using it.
If you browse through the techniques, you will see that
many of them have been developed to help with human
factors issues associated with the introduction of new
technologies – especially computer-based systems. These
new systems have been having a major impact on the
railways in recent years, and can be expected to have an
even bigger impact in the coming years. Accordingly, the
techniques described in this section are likely to have an
increasing value in the railway industry and their use will
become much more widespread than so far.
An authoritative collection of human factors techniques is
summarised in DEFSTAN 00-25, the MOD’s standard for
human factors design.
A more detailed treatment of a comprehensive collection
of these and other human factors techniques may be
found in a book by Stanton et al published in 2005. In
particular, this book describes the processes involved in
using the techniques, as well as worked examples.
Many of these techniques are also summarised in
Annex 3 of a recent major document on human factors
integration published by the Ministry of Defence’s Sea
Systems Group published in 2006.
All three of these key references are provided below.
Alternatively you might want to talk with your
organisation’s human factors specialists. If so, the accounts
of the techniques you find in this Guide may help you
have a more informed discussion with them.
Key references
1. Def Stan 00-25 (2004) Human Factors for Designers
of Systems Part 15 Issue 1, Principles and Processes. As
of May 2008, you can download this document directly
from the MOD’s DefStan website here:
www.dstan.mod.uk/data/00/025/15000100.pdf
2. Ministry of Defence (2006), MAP-01-011 Human
Factors Integration Technical Guide, Sea Systems
Group, TES-SSG-ShipDes, Defence Procurement
Agency, MOD Abbey Wood, Bristol, BS34 8JH
As of May 2008, you can download this document
directly from the HFI-DTC website here:
www.hfidtc.com/pdf/MAP-01-011.pdf
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
It is beyond the scope of this guide to describe how
to carry out all these techniques in detail. If you are
interested in a specific technique, we have provided
references where you can find out more.
Understanding Human Factors/June 08
Page 135
Part 3: Reference
Techniques
Which technique?
Technique
Input requirements
Output format
Primary users
Useful for...
Not useful if...
Applied Cognitive Task
Analysis, p139
Structured interviews
Tabulated mental skills and
demands
System designers
Understanding the mental skills and demands that a design will
require from its users
No access to task expert
Brainstorming
p140
Structured small group
discussions
Structured list
Anyone
Generating ideas from a mixed group without letting status or
seniority get in the way
Little chance of acting on results
Checklists
p141
Structured paper or electronic
form
Checked list
Anyone
Any kind of audit where a structure or systematic procedure is
needed
Situation requires explanations of
observed results
Cognitive Mapping
p142
Interviews, workshops
Graphical representation of
concepts
Anyone
Understanding the relationships between different perspectives or
concepts – often useful to organise the results of a workshop
n/a
Cognitive Walk-through
p143
Structured interviews
Video, audio or paper record of
user responses
System designers
Evaluating the usability of user interface designs
Not enough time
Communications Usage
Diagram, p143
Interviews and observations
Tabulated graphical data
System designers
Understanding the communications and collaborative
requirements of geographically separated team members
Not enough time
Critical Decision
Method, p144
Structured interviews
Annotated incident timeline
System designers
Training designers
Understanding the mental skills and demands that were required
by expert decision makers for past actual situations
No access to human factors experts
who are also skilled interviewers
Critical Incident
Technique, p145
Structured interviews
Annotated incident timeline
Incident investigators Understanding the decisions and rationale of participants in past
System designers
actual situations
No access to skilled interviewer
Design Scenario
Analysis, p145
Scenario storyboards
Video, audio or paper record of
user responses
System designers
Design team unavailable to act out
scenarios
Fault Trees
p146
Actual incident report or detailed
scenarios (for prospective events)
Graphical representation of root
causes
Incident investigators Analysing the root causes of actual or potential incidents
Risk analysts
Incident involves large complex systems
Focus Groups
p147
Structured small group
discussions
Report
System designers
Understanding user or other stakeholder requirements
Process is not planned or participants
are not representative
Groupware Task Analysis
p147
Focus groups, brainstorming,
scenario storyboards
Tabulated team task data
System designers
Understanding current team problems and migrating current team
tasks to new system designs
Not enough time
Heuristic Analysis
p148
Task-driven interview
Tabulated user response data
System designers
Rapidly obtaining user feedback about an interface design
Articulate task experts are unavailable
Hexagons
p149
Structured small group
discussions
Graphical representation
Anyone
Understanding the relationships between different perspectives or
concepts - builds on cognitive mapping by revealing gaps
n/a
Hierarchical Task
Analysis, p149
Interviews, focus groups,
observation, questionnaires etc
Graphical representation and
tabulated task data
Anyone
Understanding the detailed behavioural requirements, demands
and procedures for successful task performance
Task is largely cognitive rather than
behavioural
HEART
p151
Structured process using predefined factors and scales
Tabulated error data and
reduction strategies
System designers
Risk analysts
Quantifying predicted human error in new systems
Task expert is not available
Page 136
Proposing, evaluating and modifying design concepts without the
need for expensive simulation
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Which technique?
Technique
Input requirements
Output format
Primary users
Useful for...
Not useful if...
Human Error HAZOP
p152
Structured small group
discussions
Tabulated error data
System designers
Risk analysts
Prospectively assessing the likelihood and nature of errors in a
system under design
Mixed team of engineers, operations
staff, HF specialists unavailable
Human Reliability
Analysis, p153
Interviews, focus groups,
observation, questionnaires etc
Graphical representation and
tabulated data
System designers
Risk analysts
Analysing the probability of human error during system operation.
HRA refers to a family that includes Fault Trees, HE HAZOP etc
Varies with its specific instances such as
Fault Trees, HE HAZOP
Instantaneous SelfAssessment, p153
Operator’s on-job use of
electronic device at intervals
Workload profile across a shift
Anyone
Estimating operator workload throughout a work shift in both real
world and simulated tasks
Specific sources of workload need to
be identified
Integrated Performance
Modelling Envt, p154
Task and system specifications,
and performance requirements
Time and error forecasts
System designers
Training designers
Modelling operator (and team) tasks, workload etc in normal,
degraded and abnormal conditions for proposed complex systems
Task or human factors experts are not
available
Interface Surveys
p155
Structured survey form
Tabulated and summarised user
response data
System designers
Evaluating the physical aspects of a user interface against available
human factors criteria, guidelines and standards
No access to human factors expert
Interviews
p155
One to one discussions
Structured or tabulated report
Anyone
Collecting both qualitative and quantitative data - in which people
are asked open ended or closed questions respectively
Interviewer biases are not successfully
addressed
Keystroke Level Model
p156
Structured process applied to a
task analysis
Keyboard task timings
System designers
Estimating times for keyboard-based tasks
Tasks are not computer-based or user
performance is not serial
Layout Analysis
p157
Interviews, walk-throughs and
observation
Revised interface layout
System designers
Designing an interface based on the user’s mental model of the
way the task should be done
Users are unavailable or tasks are very
complex
Link Analysis
p157
Interviews, walk-throughs and
observation
Tabulated and graphical interface
design data
System designers
Designing an interface based on operational relationships
(frequency and importance) between different task elements
Cognitive issues outweigh the physical
relationships between task elements
Murphy Diagrams
p158
Task analysis
Structured graphical root cause
data
System designers
Risk analysts
Analysing the root causes of past and potential errors for both
single operator and team tasks
Task is large and complex
NASA Task Load Index
p159
Operator’s on-job response to
probe questions
Overall workload score
System designers
Risk analysts
Estimating operator mental workload, based on well validated
principles
n/a
Observational Analysis
p159
Video, audio or other data
recording site preparation
Video, audio or paper record of
user behaviour
Anyone
Collecting information about the visual or verbal aspects of an
actual task or scenario
Properly trained observers who are
also task experts are not available
Questionnaire
p160
Structured design of the
questionnaire form
Summary statistics
Anyone
Collecting large amounts of attitudinal data very cost-effectively for
numerical analysis and summary
Analysis of performance is required
(not analysis of what people say)
Qu. for Dist’d Assm’t of
Teams, p161
Team on-job response to
questionnaires
Summary situation awareness
scores
Anyone
Estimating the degree of mutual situational awareness of team
members
Too much time has passed between
task performance and use of technique
Qu. for User Interface
Satisfaction, p161
Pre-defined questionnaire
Global and sub-scale usability
scores
System designers
Assessing usability of user interfaces
n/a
Sequentially Timed
Events Plotting, p162
Actual incident report; interviews
Graphical timeline
Incident investigators
Identifying the contributory events leading to an incident and
facilitating discussion between investigators
Historical, time-based incident data is
unavailable
Situation Awareness for
SHAPE, p163
Operator’s on-job response to
probe questions
Summary situation awareness
scores
System designers
Assessing situation awareness
Task expert who can conduct probe
questions on-job is not available
Understanding Human Factors/June 08
Page 137
Part 3: Reference
Techniques
Which technique?
Technique
Input requirements
Output format
Primary users
Useful for...
Not useful if...
Situation Awareness
Rating Technique, p164
Operator’s on-job response to
pre-defined rating scales
Overall situation awareness score
for each team member
System designers
Training designers
Estimating the degree of situational awareness of team members
Thorough analysis of situation
awareness is required
Soft Systems
Methodology, p164
Interviews and workshops with
stakeholders
Model of the problem area and an
action plan to deal with it
System designers
Change managers
Establishing ‘rich pictures’ of complex organisational situations in
order to characterise and solve problems
No access to experienced analysts - or
the time to work with them
Software Usability Meas. Operator’s post-task response to
pre-defined questions
Inventory, p165
Global and sub-scale usability
scores
System designers
Assessing usability of user interfaces in detail
Unwilling to spend €1,000 on the
purchase price
Subjective Workload
Assess. Technique, p166
Operator’s response to predefined questions
Individual and overall workload
scores for several dimensions
System designers
Estimating multiple aspects of workload for tasks based either on
actual advanced system designs or a task analysis
Accurate detail is required for mental
workload (vs time or stress)
System Usability Scale
p167
Operator’s post-task response to
pre-defined questions
Simple overall usability score
System designers
Rapid, reliable, rough assessment of the usability of an interface
design
Detailed usability analysis is required
Syst. Hum. Err. Red. &
Pred. Approach, p167
Structured process using predefined error taxonomy
Tabulated errors and suggested
remedies
System designers
Risk analysts
Predicting human error from task analysis data, and identifying
error reduction strategies
Cognitive components of error are
critical to the task
Task and Training Reqts
Methodology, p168
Structured process based on a
task analysis
Tasks prone to skill fade, training
gaps, required skills and teamwork
Training designers
Training needs analysis for team - especially for skills prone to skill
fade
Task expert or sufficient time are not
available
Task Centred System
Design, p169
Structured process using an
existing design
Principled re-design of system
System designers
Evaluating system design concepts using user-centred principles
Team of task expert, design engineers
and system operators is not available
Team Cognitive Analysis
p170
Observation and interviews using
CDM-type structured process
Tabulated decision requirements
System designers
Training designers
Understanding the decisions and decision making skills used by
team members on a team task
Not enough time available or too
much time elapsed since task
Team Decision Reqts
Exercise, p171
Structured interviews
Tabulated decision requirements
System designers
Training designers
Understanding the thought processes of team decision makers
Two skilled interviewers who are
familiar with the technique not available
Team Workload
Assessment, p171
Team member’s on-job response
to probe questions
Workload scores for each team
member and the team as a whole
System designers
Estimating team workload, based on the NASA TLX technique
n/a
Teamworking Improv’t
Process, p172
Questionnaire and workshops
Prioritised remedies for specific
teamworking problems
Change managers
Diagnosing and identifying cost-effective remedies to problems of
teamworking within an organisation
n/a
User Trial
p173
Interviews, focus groups,
workshops, observation
Refined user requirements, training System designers
and skills recommendations
Training designers
Evaluating a new system in terms of its usability and training
requirements
n/a
Walk-through Analysis
p173
Focus groups, workshops,
interviews
Video, audio or paper record of
user responses
System designers
Evaluating early system design concepts
Task experts are not available
Why-Because Analysis
p174
Actual incident report
Annotated, graphical
representation of the incident
Incident investigators
Analysing the root causes of actual incidents and suggesting
countermeasures
Experienced incident analyst is not
available
Workshops
p175
Structured larger group
discussions
Report
Anyone
Understanding user or other stakeholder requirements and
feedback - and getting their ownership of a project
Process is not planned or participants
are not representative
Page 138
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Applied Cognitive Task Analysis
Why is it useful?
Like most other areas, work on the railways has shifted
over time from very physical work towards work that
is much more ‘mental’ (also referred to as ‘cognitive’).
People are doing less in an obviously active way such
as moving physical objects and controls. Instead they
are doing much more work that goes on largely inside
their heads: monitoring, thinking and deciding. Think, for
example, of the difference between the work of a driver
and fireman on a 1950s steam engine and the driver in
the cab of a modern Pendolino train. Or between the
work of a signaller in a traditional lever frame signal box
and that of a signaller in an IECC signalling centre.
Much of this change can, of course, be attributed to the
introduction of computer-based systems and equipment.
The designers of these systems, and their associated
jobs and training need to be able to understand and
describe the cognitive tasks they are expecting people
to undertake. But compared to traditional physical work,
mental tasks are largely hidden, and one cannot see
what the person is doing. It is to help make these tasks
visible so they can be designed for and managed that the
Applied Cognitive Task Analysis (ACTA) technique has
been developed.
What does it do?
Cognitive Task Analysis is very useful for clarifying the
thinking, decision making and judgement skills that tasks
require, so that effective equipment and interface designs,
and training, can be created. Applied Cognitive Task
Analysis (ACTA) is one of the best of its class.
Understanding Human Factors/June 08
What does it involve?
Finding out more
ACTA is a suite of interview techniques designed to
identify the thinking demands imposed by a task on an
operator. The technique is aimed at situations where the
most of the task goes on inside an operator’s head (eg
a train driver) rather than through observable behaviour
(eg a trackman).
1. Militello L.G. & Militello J.B. (2000) Applied Cognitive
Task Analysis (ACTA): A practitioner’s toolkit for
understanding cognitive task demands. In J. Annett & N.
S Stanton (Eds) Task Analysis, 90-113, Taylor & Francis
ACTA requires three kinds of interview with subject
matter experts (SMEs) to be undertaken. First, the analyst
works with one or more SME to prepare an overview of
the task, usually set out as a diagram. Second, the analyst
works with SMEs to clarify the knowledge needed by
operators to perform each part of the task revealed in
the task diagram. Third, the analyst works with the SME(s)
through specific task scenarios in order to probe specific
situations which are difficult or complex.
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
The results of these interviews are typically entered
into a tabular format, which can then be used to inform
either the design of the system or the design of training
procedures.
Who can use it?
ACTA is primarily aimed at supporting the system design
process. The success of the technique depends both on
the availability of knowledgeable and articulate SMEs,
as well as on ACTA analysts who are skilled at probing
expert knowledge. While no specialist training in cognitive
psychology is required, it is highly desirable to involve a
human factors specialist in the use of the technique.
Page 139
Part 3: Reference
Techniques
Brainstorming
Why is it useful?
Safety on the railways can never be taken for granted.
Circumstances are always changing and new ideas are
constantly needed to ensure high levels of safety are
maintained. Good ideas are not the monopoly of senior
managers. Indeed, in relation to safety, it is often the staff
at the sharp end – in the signal box, on the train, working
on the line or on the platform – who have the best ideas
on how safety can be improved. But these ideas can only
be implemented if they are first brought to light. Often
people in less senior positions are reluctant to make
suggestions for fear of being criticised or because they
think their ideas may seem foolish if they are not yet fully
worked out. Brainstorming is a way of getting ideas from
everyone out in the open so they can be developed and
assessed.
What does it do?
Brainstorming is an easy to use discussion technique
for generating ideas from a group without letting status
or seniority get in the way. See also Hexagons for a
complementary technique.
What does it involve?
A brainstorming session is a loosely structured discussion,
in which group members are encouraged to float ideas.
Although the group leader should have a clear idea of
the problem to be solved, limited guidance is given and
all participants are encouraged to put forward ideas
without the fear of criticism. For example, an operations
manager might conduct a session among engineering
supervisors and COSSes in order to generate ideas and
Page 140
possible solutions for the management of a complex T3
possession.
A brainstorming session may typically last up to an
hour or more. In the first phase, group members are
encouraged to generate ideas without criticism. However
outlandish they might seem, they are all noted without
discussion where everyone can see them eg on a flip
chart. In the second phase, the group members agree
some evaluation criteria and then review each of
the ideas, discarding any that appear to be infeasible.
Sometimes it is helpful to use a formal vote (eg show of
hands, secret ballot) to eliminate or prioritise ideas.
Finding out more
1. Rawlinson J.G. (1981) Creative thinking and
brainstorming, Gower
2. Van Gundy A.B. (1981) Techniques of structured
problem solving, van Nostrand- Reinhold, New York
The result should be a list of ideas about what is involved
in the problem, or what the possible solutions are.
Brainstorming is especially useful where everyone’s views
are equally important, but where discussions could be
dominated by individuals of higher status. By seeking a
wide range of views – initially, uncritically – over a range
of ideas, brainstorming can achieve a comprehensive view
of a problem and possible solutions.
Who can use it?
Anyone in the railway industry over wide range of
problem areas. No special skills or resources are required
other than a group of people, a quiet room for an hour
or so and a flip chart.
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Checklists
Why is it useful?
There are many working situations on the railways where
safety depends on ensuring that all necessary actions
have been completed, and often completed in the right
order. The process that a signaller and a PICOP have to
go through when taking or handing back a possession is a
good example here. Another example is the process that
depot staff should follow in checking that a train is fit to
enter service. Many safety-related incidents arise because
a step has been missed out or not completed properly.
Requiring staff to use a checklist is a straightforward way
of minimising the risk that these types of error can occur.
What does it do?
A checklist is a simple and powerful tool that is in
everyday use in a wide range of application areas.
Its primary use is to check situational status eg of a
possession or equipment.
What does it involve?
Checklists may already exist or else need to be
developed from scratch for a particular purpose.
Developing a checklist takes considerable attention and
must be based on authoritative documentation and
subject matter expert knowledge. Brainstorming and/or
workshops can help in the checklist development process.
This is to ensure that the items on the list are both
relevant and appropriate for the purpose of the check.
For example, a checklist to support a depot maintenance
schedule for a train must contain all of the right items
for that service interval without omissions. Similarly, the
list should not allow simply a tick in the box if a more
Understanding Human Factors/June 08
detailed investigative procedure is more appropriate. The
flip side of this point is also important: checklists are not
generally useful for explaining why things fail.
Checklists are useful in supporting the evaluation of
designs of new equipments or interfaces. The references
below provide some examples of evaluation checklists
that have already been developed for this purpose. If you
decide that a new checklist needs to be developed, you
should ensure that input from designers, task experts, a
range of users and subject matter experts is utilised in
the checklist design. It is vital that the checklist is reviewed
thoroughly by its target users and subject matter experts
before being deployed.
Who can use it?
Anyone across the railway industry can use a preexisting
checklist with minimal training, although they may need
to be specially qualified to make the necessary technical
judgments demanded. A more specialist group of subject
matter experts is normally needed to design and evaluate
a new checklist.
Finding out more
1. Ciavarelli A. (2002) Human Factors Checklist: An
Aircraft Accident Investigation Tool, School of Aviation
Safety, California
2. Human Engineering Ltd (2004) The Human Factors
SPAD Hazard Checklist, Issue 2 HEL/RSSB/041123/
RTB02, Network Rail
3. Kirwan B. & Ainsworth L.K. (eds) (1992) A guide to
Task Analysis, Taylor & Francis
4. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
5. Ravden S.J. & Johnson G.I. (1989) Evaluating Usability of
Human-Computer Interfaces: A practical method, Ellis
Horwood, Chichester
6. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Page 141
Part 3: Reference
Techniques
Cognitive Mapping
Why is it useful?
Nearly every initiative to develop some feature of the
railways involves groups of people working together,
whether it is a group of designers working on a
new system such as GSM-R, or a group of managers
considering changes to the Rule Book. These groups often
involve people from varied occupational backgrounds
and experience. A major obstacle to progress is that
all the participants see the topic they are addressing in
totally different ways. Worse, they may be unaware that
others see the issue so differently. This can often lead
to misunderstanding, confusion – and even argument.
What is needed is a way building up a way of visual
representation of the topic or situation, so that different
assumptions and viewpoints can become apparent, and
eventually an agreed picture can be created. Cognitive
mapping is one way of achieving this.
What does it do?
Cognitive mapping is a group activity that is useful for
focusing the activity of discussion – eg in a workshop –
where a key aim is to create a shared understanding
of the different perspectives and requirements of the
participants.
What does it involve?
Cognitive mapping produces a kind of map of how
the participants see a situation or problem area. The
map (also known as an influence diagram) is made up
of concepts (represented as words or short phrases)
and arrows that point from means to ends (ie from
contributing factors to goals). A minus sign shows an
Page 142
inverse relationship between two concepts ie the more
of one concept, the less of the other.
Example of a cognitive map (influence diagram)
Finding out more
1. Eden C. (1989) Using cognitive mapping for strategic
options development and analysis (SODA), in
Rosenhead J. (Ed.), Rational analysis for a problematic
world, 21-70, John Wiley
2. iSee Systems STELLA™ and iTHINK™ www.
iseesystems.com/ (as of May 2008)
This example shows the positive and negative influences
held by a group of people on the rate by which
organisational innovations (eg new IT systems) are likely
to be adopted by staff. The diagram shows their belief
for example that the more complex the system is, the
less rapidly it will be adopted. Similarly, the greater the
relative advantage of the new system, the more rapidly
it will be adopted. Interesting and more complicated
patterns of influence can be established when lines
of influence can be drawn between many nodes. It is
possible to calculate the effects of the influences on
each other by representing the influence diagram in
affordable computational systems thinking software such
as STELLA™ or iTHINK™ (see References).
Who can use it?
Anyone capable of planning and running a workshop,
although generating influence diagrams takes a little
practice to get right.
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Cognitive Walk-through
Why is it useful?
Technology is providing the railways with many new
forms of equipment and system. These can bring
important benefits, but often, the new system is more
complex than the one it replaces. This complexity may
not be immediately obvious to the user, eg the IECC
screen. The means by which the operator uses the
equipment, how they communicate with it and how
it communicates with them, is known as the ‘humanmachine interface’ or ‘human-computer interface’ (HCI).
The design of this interface is vital to ensuring safe and
reliable performance. Before any design is finalised its
usability must be assessed. One technique for conducting
a usability assessment is the Cognitive Walk-through.
What does it do?
Cognitive Walk-through is a structured observation
technique that helps an analyst evaluate the usability
of an interface. The technique is a development of the
traditional design walk-through method.
What does it involve?
The technique’s emphasis is on examining the ease
with which a new interface can be learned. The analyst
makes this examination by applying a set of criteria that
are concerned with the cognitive (thinking) processes
required by the user to perform the task. Examples of
these criteria include how obvious it is for the correct
choice to be made, how clearly labels are linked to
their associated actions, and how easy it is for the user
to understand what they have done. The analyst ‘walks
through’ each action that a user needs to take to carry
Understanding Human Factors/June 08
out a task using prototype interfaces for driving a new
type of train or using a new ticket machine. Each action
is scrutinised in terms of technique’s pre-defined criteria
in order to estimate the impact made by the interface on
the user’s goals and actions.
Who can use it?
Cognitive Walk-through is primarily aimed at supporting
the system design process. Its success depends both on
the availability of knowledgeable and articulate task or
subject matter experts, as well as on analysts who are
skilled at applying the pre-defined criteria. It can also
be quite time-intensive. While no specialist training in
cognitive psychology is required, it is highly desirable to
involve a human factors specialist.
Finding out more
1. Polson P.G. Lewis C. Rieman J. & Wharton C. (1992)
Cognitive walk-throughs: a method for theory based
evaluation of user interfaces. Int Jnl of Man-Machine
Studies, 36 pp741-773
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
4. Wharton C. Rieman J. Lewis C. & Polson P. (1994)
The Cognitive Walk-through Method: A Practitioners
Guide, John Wiley
Communications Usage Diagram
Why is it useful?
Safety on the railways depends on teamwork. But many
teams on the railways are not what people normally
think of as teams. This is because teams on the railways
often involve people working in different places and often
only working together for a short time. For example,
a train driver and a signaller may only be working
together for a few minutes. During that time they share
the same goals and they are dependent on each other.
These arrangements are known as distributed teams. In
such short-lived, physically separated teams the quality
of communications is of the utmost importance. The
Communications Usage Diagram can be a valuable aid for
designers, trainers and managers responsible for ensuring
that distributed teams can communicate when and how
they need to.
What does it do?
A Communications Usage Diagram (CUD) is a technique
for analysing and describing the communication aspects of
distant teamworking (page 103).
What does it involve?
Creating a Communications Usage Diagram (CUD)
involves specifying what is communicated between the
geographical locations of each member or group that
comprises a team. For each type of communication, the
technology used is also specified, together with the pros
and cons of that medium. The technique could potentially
be applied in any domain involving communication
or collaboration. For example, a CUD could be used
to document the problems of communications and
Page 143
Part 3: Reference
Techniques
supporting technologies between signallers, trackworkers,
drivers and control at a particular location in order to
determine the requirements for new communication
procedures or technologies.
Who can use it?
Anyone involved in analysing communications problems
and the potential for improvements. However, sufficient
time should be allowed for the initial data collection,
which can involve many interviews and observations.
Finding out more
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
3. Watts L.A. & Monk A.F. (2000) Reasoning about tasks,
activities and technology to support collaboration.
In J. Annett & N. Stanton (Eds) Task Analysis, Taylor &
Francis
Page 144
Critical Decision Method
Why is it useful?
In most situations signallers, train drivers, engineering
supervisors, and other people working in safety critical
roles on the railways can act in accordance with the Rule
Book. But there are inevitably many situations in which
the Rules do not apply, are conflicting or are unclear.
This means rail workers have to make decisions which
are often difficult and may have serious consequences if
wrong. The Critical Decision Method is a means to help
designers and planners understand these decisions in
detail. It is a later version of the Critical Incident Technique.
What does it do?
The Critical Decision Method (CDM) is a relatively
powerful, easy-to-use structured interview technique
for revealing the thought processes of expert decision
makers. It is useful for understanding task requirements
for both system and training design.
What does it involve?
CDM works by means of a set of pre-defined probe
questions that an analyst asks an expert about their
decision-making. For example, an analyst could discover
what cues the expert was using by asking “How did you
know that you needed to make the decision?” Similarly, the
expert’s situational awareness could be probed with a
question such as “What information did you have available
to you at the time of the decision?” Video can be a useful
means of ensuring that information generated by the
probe questions is not missed, although video analysis
can be time consuming. The information yielded can then
be used to inform the development of new systems
and/or associated training systems. CDM is useful at a
very early stage in the design of a new operational or
training system. For example, it could be used to infer the
requirements for a new cab design from the actions of
expert drivers using existing cabs.
Who can use it?
CDM requires at least two interviewers, preferably
human factors specialists, who need to be reasonably
skilled in the technique in order to get the most out of it.
Finding out more
1. Flanagan J.C. (1954) The Critical Incident Technique.
Psychological Bulletin, 51, 327-358
2. Klein G.A. Calderwood R. & MacGregor D. (1989)
Critical Decision Method for Eliciting Knowledge. IEEE
Transactions on Systems, Man and Cybernetics, 19(3),
462-472
3. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
4. O’Hare D. Wiggins M. Williams A. & Wong W. (2000)
Cognitive task analyses for decision centred design and
training. In J. Annett & N. Stanton (eds) Task Analysis, pp
170-190, Taylor & Francis
5. Stanton N., Salmon P., Walker G., Baber C., & Jenkins
D., (2005) Human Factors Methods (2005) Human
Factors Methods: A Practical Guide for Engineering
and Design, ISBN: 0754646610, Ashgate Publishing
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Critical Incident Technique
Why is it useful?
In many situations on the railways it is important to know
why people made the decisions that they did. This may be
necessary to understand how a particular incident was
caused, or it may be necessary when developing systems
or procedures intended to help railway staff in safety
critical roles to make better decisions. For example, the
increased use of computer-based systems can not only
provide staff with better quality information, such systems
can also actively help staff predict how events will unfold
in the future. This technique enables an analyst to explore
in a systematic way how railway personnel have made
particular incidents in critical situations in the past.
However, like all retrospective techniques, it is subject to
the unreliability of human memory and the interviewee’s
tendency to distort history by trying to make sense of
things after an event. It is sometimes difficult to build
a good timeline unless other corroborating data are
available. A key value of CIT is that it can reveal poor
design of both equipment and processes.
Who can use it?
CIT is a very useful and inexpensive investigative
technique for someone already trained in interviewing
skills. It is not necessary to be a human factors specialist
to use the technique, although access to this specialism is
of undoubted assistance in interpreting the results – and
possibly guiding some of the questions.
What does it do?
Finding out more
Critical Incident Technique (CIT) is an interview technique
designed to help people reconstruct their roles, decisions
and reasons for acting in the way they did, during the
critical phases of a past, actual incident. CIT is an older
techniques that has been developed more recently into
the Critical Decision Method (CDM).
1. Flanagan J.C. (1954) The Critical Incident Technique.
Psychological Bulletin, 51, 327-358
What does it involve?
CIT involves an analyst using a set of pre-defined probes
within an interview convened for the purpose. The
probes are a forerunner of the much more extensive
probes used in CDM. They are simply aimed at eliciting
information about the circumstances leading up to
an incident, the critical actions or inactions, the results
and other actions or inactions that might have made
a difference. Today, such questions form the basis of
debriefing railway staff after an incident affecting safety.
Understanding Human Factors/June 08
2. Klein G.A. Calderwood R. & MacGregor D. (1989)
Critical Decision Method for Eliciting Knowledge. IEEE
Transactions on Systems, Man and Cybernetics, 19(3),
462-472
3. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
4. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Design Scenario Analysis
Why is it useful?
Like any business, the railways need to constantly
innovate to improve performance and safety. Being able
to imagine not only new design concepts but also their
use in a variety of settings is a powerful way to innovate.
Design scenario analysis provides the means to explore
new concepts in a wide range of new contexts.
What does it do?
Design scenario analysis is a technique that design teams
can use to help generate, review and refine design
concepts.
What does it involve?
Design scenario analysis works best with several - even
many - members of a design team present. Together,
they generate a new concept and set out a storyboard
for it within some operational context which they also
generate for the purpose. As they probe and challenge
it, new parts of the ‘story’ are added and changes made
to the design as a result. The techniques is not limited to
equipment design. For example, design scenario analysis
would be just as suited to the design of new procedures
for the rule book as it would be for the design of catering
trolleys for on-train use or new signalling panels. While it
may sometimes be difficult to get all of the design team
together in one place, the technique itself is inexpensive,
flexible, highly creative (and enjoyable!), often leading to
innovative results.
Page 145
Part 3: Reference
Techniques
Who can use it?
Design teams.
Finding out more
1. Go K. & Carroll J.M. (2003) Scenario-Based Task
Analysis. In D. Diaper & N. Stanton (Eds) The
Handbook of Task Analysis for Human-Computer
Interaction, Lawrence Erlbaum Associates, London
2. Verplank B. Fulton J. Black A. & Moggridge B. (1993)
Observation and Invention – Use of scenarios in
interaction design. Tutorial notes for InterCHI 93.
Amsterdam
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Fault Trees
Why is it useful?
Incidents on the railways typically involve some
component failing or a person making an error. Many
engineers and designers on the railways will be familiar
with fault trees for describing how an incident was caused
or for assessing the likelihood of a physical component or
sub-system going wrong at some future point. Fault trees
can also be used to examine how a procedure or other
human action has gone wrong or could go wrong. For
example, Rule Book designers could use the technique
to discover errors that might occur when a signaller and
COSS are arranging T12 protection, and estimating the
probability of these errors occurring.
What does it do?
Fault trees are a relatively simple but effective root cause
analysis technique for understanding and sometimes
predicting multiple causes of failure events.
What does it involve?
The analyst creates a tree-like diagram to represent
hardware failures and human errors. The contributing
causes are linked together by AND and OR relations and
each contributing cause is further analysed in terms of its
own contributing causes. Care needs to be taken to keep
the analysis going until the root causes (rather than the
preceding causes) are properly documented.
Who can use it?
Anyone faced with the task of analysing the actual or
likely causes of failure within an operational system. Little
training is required, although a logical, systematic approach
helps a lot. While the analysis of simple incidents generally
requires modest amounts of time, more complex incident
analyses can be quite time consuming.
Finding out more
1. Kirwan B. & Ainsworth L.K. (1992) A Guide to Task
Analysis, Taylor & Francis
2. Kirwan B. (1994) A Guide to Practical Human
Reliability Assessment, Taylor & Francis
3. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
4. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
The original use of this technique was to identify the root
causes of actual incidents. However, it is also possible to
use it throughout design in a predictive way, in order to
try to eradicate failures before they arise.
Page 146
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Focus Groups
Why is it useful?
Focus groups have become mainly known for their use
in the worlds of market research and politics. Market
researchers use them to gauge consumer reaction to
products and to work out how they can be presented
more attractively. Political parties use focus groups
to assess public opinion in relation to some issue of
interest. If participants are correctly chosen, the right
questions asked, and the right procedures followed, it
has been found that a focus group with a small number
of participants can be as accurate as a more traditional
survey involving large numbers of people. There are many
potential uses of focus groups in the railways. As well
as finding out the views of, say, members of the public
or staff members, focus groups can be used by more
specialist groups such as trainers or system designers to
establish how users will respond to a specific issue.
What does it do?
A focus group is a structured discussion that uses a small
group of stakeholders to discuss a particular set of predefined topics. It is a smaller and more highly constrained
version of a workshop.
system or to understand usability or workload issues of
a new prototype; training designers might use them to
modify a training method among a group of trainers.
Who can use it?
Experienced facilitators can enable focus groups to
be a very useful, trustworthy and economic source of
stakeholder information.
Finding out more
1. Hyponen H. (1999) Focus Groups. In H. A. Williams,
J. Bound & R. Coleman (eds) The Methods Lab: User
Research for Design. Design for Ageing Network
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Groupware Task Analysis
Why is it useful?
Groupware refers to a class of technical systems designed
to improve the work of any collection of people who
have to work together to achieve some defined set of
goals. Groupware supports the more rapid and efficient
sharing of information and passing of documents so
that different individuals can each make their particular
contribution to the on-going work. So far groupware
has been mainly used for traditional office tasks, but
there is immense scope for wider exploitation across
the railways. For example, groupware offers the potential
for much more reliable passing of information between
staff involved in planning engineering works, the staff
who will actually carry out the work, the signallers who
must arrange protection with the PICOP, and the TOC
staff who may be affected by the engineering works.
Groupware systems are inevitably complex and careful
analysis is required to ensure they will work correctly.
What does it do?
Groupware Task Analysis (GTA) is a technique for team
task analysis. It is designed to migrate current team tasks
to new system designs.
What does it involve?
What does it involve?
Focus groups normally involve a group of up to 8 people
and 1 to 2 facilitators, and typically last for 2 to 4 hours.
Typically, the facilitators encourage discussion of each
topic by describing situations or scenarios and asking
probe questions designed to reveal participant views and
their rationale. In a railway setting, system designers might
use them to elicit user requirements for a proposed
The GTA analyst uses structured interviews and
observation of team activities within current systems, The
results are used to contribute to the design of new teambased systems. GTA provides a framework for a two part
analysis. In the first part, the analyst describes the current
system, eg signalling. This is done in order to ensure that
the design team understands the current workflows
Understanding Human Factors/June 08
Page 147
Part 3: Reference
Techniques
and communication patterns between signallers, drivers,
control etc. In the second part, the analyst re-designs the
current team task using new technologies to solve any
problems and requirements presented by the current
situation. The result is a representation of a proposed
system together will all its assumptions for the future
scenario(s) in which it will operate.
Who can use it?
Design teams – but it should be noted that the technique
is time-intensive and resource-hungry.
Finding out more
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
3. Van Welie M. & Van Der Veer G. (2003) Groupware
Task Analysis. In E. Hollnagel (Ed) Handbook of
Cognitive Task Design. Pp 447 – 477. Lawrence
Erlbaum Associates
Page 148
Heuristic Analysis
Why is it useful?
New equipment and systems – especially those that are
computer-based – often provide many more facilities
and features than the old equipment or system. The
price paid for this new capability is that the interface (the
means whereby the user interacts with the equipment
or system) becomes much more complex and therefore
more difficult. There are numerous techniques described
elsewhere in Part 3 that provide thorough but timeconsuming means of assessing user interfaces. The value
of Heuristic Analysis is that it is quick and easy to use,
although limited in certain respects.
What does it do?
Heuristic analysis is a rapid, simple, subjective technique
for recording opinion about (eg) a task interface.
What does it involve?
Who can use it?
Anyone can use heuristic analysis, but its results will only
be as good as the analyst’s ability and imagination to spot
the negative aspects of the task during the walk-through.
Finding out more
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
2. Nielsen J. & Molich R. (1990) Heuristic evaluation
of user interfaces. In J. C Chew & J. Whiteside (eds),
Empowering people: CHI 90 Conference Proceedings
(pp. 249 – 256) Monterey, CA: ACM Press
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
The analyst walks the user through their task as
systematically as possible and tries to spot the
opportunities for error and other design weaknesses
as they go. The technique can be re-used throughout
the design process. The walk-through can be structured
by the analyst in any way they choose, and any such
structuring is likely to improve the quality of the results.
Note, though, that any undesirable effects arising from
complex interactions with other tasks or events are
unlikely to arise or be spotted.
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Hexagons
Finding out more
Hierarchical Task Analysis
Why is it useful?
1. Hodgson A.M. (1992) Hexagons for systems thinking,
European Journal of Operational Research, 59,
220-230
Why is it useful?
There are many situations on the railways where a
group of people need to come up with good ideas.
Brainstorming is perhaps the most commonly used
technique to generate lots of ideas. While writing
ideas that the group comes up with on a flipchart or
whiteboard will often work well, what is even better is
to be able to move ideas around, so similar ideas can
be grouped together. Being able to see what ideas have
in common or how they differ, can often enable the
group to develop the ideas further. Hexagons is a simple
technique for helping with this process.
What does it do?
Hexagons is a simple technique which can be used to
support brainstorming. It helps with the structuring of
emergent ideas and the evaluation of their completeness.
What does it involve?
Hexagons involves a workshop session in which issues
written on magnetic hexagons (alternatively, coloured
post-it notes) can be grouped and classified. It is a way for
a group to get a comprehensive picture of the problem
they are tackling and it pushes the group into thinking
about the gaps in the discussion so far.
Who can use it?
Anyone in the railway industry over a wide range of
problem areas. If a magnetic whiteboard and magnets are
not available, post-it notes are a good substitute.
Understanding Human Factors/June 08
There are many situations in designing equipments,
systems, training and jobs on the railways where it
is necessary to have a clear understanding of what
the various tasks entail. One of the most widely used
techniques for doing this is hierarchical task analysis
(HTA). HTA is based on the fact that most tasks can be
decomposed into a hierarchy of goals. For example, a
signaller’s main task is to ensure the safe movement of
trains. To achieve this goal he must carry out a number
of sub-tasks: set routes, manage the taking of possessions,
monitor level crossings, and so on. Each of these sub-tasks
can be further broken down still further into smaller tasks
(eg communicating with PICOP, protecting signals, placing
reminders, completing forms). Tasks, sub-tasks and so on
are progressively analysed until all the actions that form
the fundamental building blocks have been identified.
These basic acts can then form the basis of equipment
design, training material development, and so on. HTA
is often performed as a first step in many of the other
techniques described here.
What does it do?
Hierarchical Task Analysis (HTA) is a systematic and
well used method of task analysis (page 47) for
comprehensively representing tasks as a hierarchical
structure of goals and sub-goals. It works especially well
for tasks that emphasise what operators must do rather
than what they must think and the way they should think
it.
Page 149
Part 3: Reference
Techniques
What does it involve?
HTA involves representing the task of interest as a treelike structure of goals, operations and plans.
• Goals – These provide the purpose for carrying out
the task - or some piece of it
• Operations – These are all the observable activities
that must be carried out by the operator in order to
achieve the goal
• Plans – These are the conditions for, and sequences by
which, the operator carries out the operations.
Carrying out a HTA for a large or complicated task like
driving a train or operating a signaller’s NX panel can be
quite time-consuming. While HTA affords many insights
into a task, a key problem is that even when all the
operations have been specified, it is often by no means
obvious what should be done about the results – in
terms of risk detection, error prevention or training –
without a lot of further analysis.
Who can use it?
Anyone capable of applying a logical and systematic
approach. While it is not necessary to be a human
factors specialist to use the technique, such specialists are
usually essential in interpreting and applying the results
of the analysis eg via interventions in training, system or
job design. In view of this it is recommended to involve
human factors people on the analysis team from the very
beginning,
Page 150
Finding out more
1. Ainsworth L.K. & Marshall E. (1998) Issues of quality
and practicality in task analysis: preliminary results from
two surveys. Ergonomics 41(11), 1604-1617. Reprinted
in J. Annett & N.A. Stanton (2000) op.cit. 79-89
2. Annett J. Duncan K.D. Stammers R.B. & Gray M. (1971)
Task Analysis. HMSO
3. Annett J. Cunningham D.J. & Mathias-Jones P. (2000) A
method for measuring team skills. Ergonomics, 43(8),
1076-1094
4. Annett J. Duncan K.D. Stammers R.B. & Gray M.J.
(1971) Task Analysis. Training Information No. 6. HMSO
5. Annett J. & Stanton N.A. (2000) Task Analysis, Taylor &
Francis
6. Baber, C. & Stanton N.A. (1994) Task analysis for error
identification. Ergonomics 37, 1923-1941
7. Baber C. & Stanton N.A. (1996) Human error
identification techniques applied to public technology:
predictions compared with observed use. Applied
Ergonomics 27, 119-131
10.Lim, K.Y. & Long, J. (1994) The MUSE Method for
Usability Engineering. Cambridge University Press
11.MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
12.Ormerod, T.C. Richardson, J. & Shepherd, A. (1998)
Enhancing the usability of a task analysis method:
A notation and environment for requirements.
Ergonomics 41(11), 1642-1663. Reprinted in Annett &
Stanton (2000) op.cit. 114-135
13.Patrick, J. Gregov, A. & Halliday, P. (2000) Analysing and
training task analysis. Instructional Science, 28(4), 51-79
14.Shepherd, A. (2002) Hierarchical Task Analysis, Taylor &
Francis
15.Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
8. Johnson P. Diaper D. & Long J. (1984) Tasks, skills
and knowledge: Task analysis for knowledge-based
descriptions. In B. Shackel (Ed.) Interact ‘84 - First
IFIP Conference on Human-Computer Interaction.
Amsterdam: Elsevier, 23-27
9. Kirwan B. & Ainsworth L.K. (1992) A Guide to Task
Analysis, Taylor & Francis
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Human Error Assessment and
Reduction Technique
Why is it useful?
Many people on the railways fulfil safety-critical roles –
drivers, signallers, track workers, crossing keepers, station
staff, and others. Safety on the railways depends critically
upon identifying the kinds of errors these staff (and
sometimes other groups, such as passengers) can make.
The Human Error and Assessment Technique (HEART)
is a relatively simple technique that not only helps railway
personnel identify types of error, but also to assess the
probability of each type of error occurring. This analysis
provides a basis for then looking at ways of reducing the
risks these errors occurring.
What does it do?
Human Error Assessment and Reduction Technique
(HEART) is a technique for estimating the type and
amount of human error likely to occur within a system,
and helping to avert it.
What does it involve?
HEART is easy to understand and quick to apply. It
requires the definition of representative task scenarios
and then provides a straightforward process for assessing
these scenarios with respect to the likelihood of human
error. For example, a depot manager might examine all
of the tasks that need to be undertaken by a fitter in
carrying out a maintenance programme for a new train.
The assessment process involves making judgements
about human reliability in carrying out the main elements
of the task by choosing between a number of preUnderstanding Human Factors/June 08
defined factors such as “complex skill requiring high
level of comprehension and skill” or “fairly simple task
performed rapidly or given scant attention”. Each statement
is associated with a pre-defined weighting score. At the
end, all of the weighting scores are multiplied to produce
an overall score, which helps inform the identification
remedial measures. These might include interventions
which involve re-designing the maintenance schedule
or re-training for it; improving the fitter’s perception of
risk; or assigning more experienced fitters etc. Helpfully,
the remedial measures can themselves be calculated to
determine their likely contribution to the reduction of the
calculated probability of human error. The technique has
been specially adapted for use in the railway industry.
Who can use it?
Anyone who can apply a logical, systematic approach: little
training is required, although a human factors specialist
is highly desirable as part of the training process, and a
subject matter expert is needed to define the scenario(s)
and assess the tasks within the scenario(s) against the
HEART framework of statements.
Finding out more
1. Kirwan B. (1996) The validation of 3 Human Reliability
Quantification techniques – THERP, HEART and JHEDI:
Part 1 – technique descriptions and validation issues,
Applied Ergonomics, Vol 27, 6, 359-373
2. Kirwan B. (1997) The validation of 3 Human Reliability
Quantification techniques – THERP, HEART and
JHEDI: Part 2 – Results of validation exercise, Applied
Ergonomics, Vol 28, 1, 17-25
3. Kirwan B. (1997) The validation of 3 Human Reliability
Quantification techniques – THERP, HEART and
JHEDI: Part 3 – Practical aspects of the usage of the
techniques, Applied Ergonomics, Vol 28, 1, 27-39
4. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
5. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
6. Williams J.C. (1986) HEART – a proposed method for
assessing and reducing human error. In 9th Advances in
Reliability Technology Symposium, Uni of Bradford
Page 151
Part 3: Reference
Techniques
Human Error HAZOP
Why is it useful?
The concept of HAZOP (Hazard and Operability) has
been widely used on the railways for some time. Human
Error HAZOP is a particular form of HAZOP that
focuses on the risks to safety posed by errors made by
people. Human Error HAZOP thus has similarities to
HEART but is more comprehensive and detailed, although
also more demanding in time and effort.
What does it do?
Human Error HAZOP (Hazard and Operability) is a
comprehensive and well accepted technique for assessing
the likelihood and nature of human errors arising from
fairly complex plant or plant processes at an advanced
stage of their design.
What does it involve?
The Human Error HAZOP procedure needs to be
applied by a team of design, engineering, operational and
human factors personnel, specially assembled for the
purpose, and led by someone who has had experience
with the technique. Such a team might be assembled to
review the way a new maintenance delivery depot will be
operated. The team agrees a set of guide words (such as
‘sooner than’; ‘later than’, ‘not done’) that they then use to
evaluate depot operations that have been systematically
described beforehand – often as a hierarchical structure
using a technique such as HTA. In our example, for
each section of the depot or depot process, potential
operational deviations are considered using the guide
words to identify weak points and recommend ways of
strengthening them.
Page 152
The technique is apparently exhaustive, but it should be
noted that even a team-based assessment will not be
completely sensitive to the way in which the ‘holes in the
Swiss cheese’ (as described in Why do accidents happen?
on page 18) can suddenly line up.
Who can use it?
The Human Error HAZOP technique is resource-hungry.
It requires a team of specially appointed disciplines
to carry it out – including a human factors specialist.
Crucially, the team leader needs to be experienced in the
technique for it to be successfully applied in new situation.
It is also time-intensive to carry out, typically requiring
several weeks of team effort.
Finding out more
1. Kennedy R. & Kirwan B. (1998) Development of a
Hazard and Operability-based method for identifying
safety management vulnerabilities in high risk systems,
Safety Science, Vol. 30, 249-274
2. Kirwan B. & Ainsworth L.K. (1992) A guide to Task
Analysis, Taylor & Francis
3. Kirwan B. (1992) Human error identification in human
reliability assessment. Part 1: Overview of approaches.
Applied Ergonomics Vol. 23(5), 299 – 318
4. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
5. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
6. Swann C.D. & Preston M.L. (1995) Twenty-five years
of HAZOPs. Journal of Loss Prevention in the Process
Industries, Vol. 8, Issue 6, 1995, 349-353
7. Whalley (1988) Minimising the cause of human error.
In B. Kirwan & L. K. Ainsworth (eds.) A Guide to Task
Analysis. Taylor & Francis
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Human Reliability Analysis
Why is it useful?
Although automation is an increasing feature of railway
operations (eg ATPS, ATWS, ARS), safety on the railways
still depends heavily on the reliability of the people who
work on the trains, in signal boxes, on platforms, and on
the track. Any initiatives to improve current safety or to
raise safety levels in the future must take account of how
reliably staff can carry out the tasks for which they remain
responsible. A variety of techniques – collectively known
as Human Reliability Analysis – are now available to help
human factors and safety specialists assess how reliably
humans can operate in a given situation.
What does it do?
Human reliability analysis (HRA) refers to a family of
techniques, all of which are aimed at identifying the
probability of different types of human error during
system operation. Such information then forms the basis
for system re-design and other error reduction measures.
What does it involve?
Several techniques are available for human reliability
analysis. These techniques include Fault Trees, HEART and
HE HAZOP. The techniques vary in terms of the skills
and knowledge they require and the time they take to
carry out. However, all of them require some form of task
analysis (page 47) to be carried out first so that the tasks
and their associated demands are clarified.
Generally speaking, HRA works by associating task
inputs, actions and outputs to categories of error. The
probability of errors occurring is calculated either by
Understanding Human Factors/June 08
generating a fault tree or through the use of principles
based on actual data, if available. The overall probability
of error is calculated by multiplying the individual task
results together. If the final result reveals a potential
problem, then the design or training assumptions can be
changed and the analysis re-run. If Tasks and system design
components associated with unacceptable potential error
rates may need to be subject to more detailed modelling.
Who can use it?
Varies with the specific technique, although human factors
specialists should generally be involved in order to ensure
the technique is used effectively and its results interpreted
accurately. RSSB is currently developing a rail specific
HRA technique. This is initially driver based, but will be
extended to other operational groups in due course.
Finding out more
1. Kirwan B. & Ainsworth, L.K. (1992) A Guide to Task
Analysis, Taylor & Francis
2. Kirwan B. (1994) A Guide to Practical Human
Reliability Assessment, Taylor & Francis
3. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
4. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Instantaneous Self-Assessment
(of workload)
Why is it useful?
There are many situations on the railways where it is
necessary to assess the workload of a given member
of staff. For example, as a result of re-signalling it may
be proposed to move a two man team (signaller and
crossing keeper) from a mechanical signal box to an area
signalling centre, and as part of this move to integrate
both old jobs into a single new job. The question then
is can one person now perform the tasks previously
carried out by two while maintaining safety levels? This
is a problem for workload assessment. Instantaneous
Self Assessment of Workload is one of a number of
techniques that can be used to make an estimate of the
level of workload associated with a particular job.
What does it do?
Instantaneous self-assessment (ISA) of workload (page
125) is a very cheap and reasonably effective way of
estimating patterns of operator workload throughout a
work shift on real-world, as well as simulated, tasks.
What does it involve?
ISA involves operators (eg signallers at an NX panel or
control room operators) briefly pausing in their tasks
every few minutes in order to rate their workload
on a simple scale of 1 (low) to 5 (high). Very often,
arrangements can be made for participants to rate
themselves using a simple keypad that can be made to
flash whenever it is time to provide a rating. More simply,
the analyst can simply obtain ratings verbally. When the
Page 153
Part 3: Reference
Techniques
shift is completed, the collected data can be graphically
summarised to show the workload peaks and troughs.
Who can use it?
Anyone, although a human factors specialist is of benefit
in order to ensure its correct use, and to assist in results
interpretation and options for action.
Finding out more
1. Kirwan B. Evans A. Donohoe L. Kilner A. Lamoureux T.
Atkinson T. & MacKendrick H. (1997) Human Factors
in the ATM System Design Life Cycle. FAA/Eurocontrol
ATM R&D Seminar, Paris, France
http://atm-seminar-97.eurocontrol.fr/ (as of May 2008)
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Integrated Performance
Modelling Environment
Why is it useful?
In changing working arrangements or developing new
systems it is often desirable to assess how well people
can perform in a range of situations eg can they work
quickly, reliably and safely enough? Human performance
is influenced by many different factors. Factors such as
time on shift, workload, and individual experience are
just a few of the important factors affecting work on the
railways. Taking all of these into account when making
estimates can often be difficult. Increasingly, computer
models are available which enable a range of factors
to be represented and which can be used to explore
human performance rapidly and reliably. The Integrated
Performance Modelling Environment (IPME) was originally
developed to assess human performance in highly
demanding military situations, but is also relevant for use
by railway system designers and others.
What does it do?
IPME is a sophisticated software-based environment for
modelling designs for prospective systems comprising
teams of people and technology. It is a commercial
product and probably the best of its class in terms of
usability (page 28), validity and reliability.
speed, error rates and accuracy. In the railway industry,
IPME might be used to model a new design for an
operations control room, complete with human team,
computers, displays and communications equipment.
Notably, IPME can be made to connect with other
simulations through industry standard protocols. This
means that the new design for a control room could
be connected to manned NX panel simulators in order
to increase functionality and thereby, the scope of the
evaluation.
When the simulation is run, the task network uses the
underlying micro-models to determine what happens.
The performance of the network is also influenced by
the way the environment model, the team model and the
performance shaping functions are set up. The simulation
can be run thousands of times in a short time in order to
build up a set of results for a whole simulated population
of operators.
Who can use it?
Analysts need training in IPME before they can use it and
the user team will need to include both a subject matter
expert and a human factors specialist.
Finding out more
1. Micro Analysis & Design Inc, Boulder Colorado, www.
maad.com/index.pl/ipme (as of May 2008)
What does it involve?
The Integrated Performance Modelling Environment
(IPME) uses simulation of the operator(s), the task, the
task environment and the prospective system design to
answer questions about human performance in terms of
Page 154
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Interface Surveys
Why is it useful?
The usability of facilities provided to train drivers,
signallers and other staff is a key factor – not only in
safety, but also in job satisfaction, stress levels, amount of
training needed, and so on. Assessing the quality of a new
user interface or one under development can be a time
consuming activity. Interface Surveys offer one reasonably
quick and simple means of gaining a representative
sample of data about user reactions to a system.
What does it do?
Interface surveys are a straightforward approach to
detecting a wide range of human-machine interface
design flaws in existing systems or advanced prototypes.
What does it involve?
The technique involves surveying one or more aspects
of the human-machine design, which needs to exist as
a working operational or prototype system. Examples
include advanced prototypes for a new signalling panel or
train cab design. The aspects to be reviewed might include
areas such as controls and displays, coding conventions
(eg colours and symbols, and environmental factors such
as noise, illumination, temperature and humidity levels). A
further, often fruitful, aspect for investigation includes an
examination of what modifications operators have made
to similar existing systems in order to overcome design
flaws eg the use of notes stuck on displays and self-made
aids. For each survey deemed relevant, the analyst needs
to produce a survey data collection form, drawing upon
available human factors criteria, guidelines and standards,
as well as known sources of interface problems.
Understanding Human Factors/June 08
Who can use it?
Human factors specialists are needed to design the
survey forms, which can then be used by a much wider
range of personnel.
Finding out more
1. Kirwan B. & Ainsworth L.K. (1992) A guide to Task
Analysis, Taylor & Francis
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Interviews
Why is it useful?
Interviews are already used for many purposes on the
railway, eg in recruitment, performance appraisal and
incident investigation. Many interviews ‘just happen’. But
to be truly effective an interview needs to be properly
planned and structured. The objectives of the interview
need to be clear and the questions need to be chosen
to best achieve the objectives. The interviewer and the
interviewee need to have prepared themselves for the
interview. Good interviewing needs to be seen as a
technique that must be learned.
What does it do?
The interview is a well-used one-to-one technique for
gathering, probing and confirming subjectively reported
information in a very wide range of application areas.
What does it involve?
Interviews need to be well structured, with considerable
prior thought being given to their objectives and logic.
Interviews may be qualitative – in which people are asked
open-ended questions designed to get at their opinions,
beliefs and values; or they may be quantitative – in which
people are asked closed questions designed to elicit yes/
no or rating scale type responses.
In most cases, the interview design should be piloted
and fine-tuned before being carried out for real. The
interviewer should be clear about how they are going to
record the information that will be generated. At the start
of each interview, the interviewer should also clarify for
the interviewee what use will be made of the information
Page 155
Part 3: Reference
Techniques
and how far it will be attributable. When conducting
an interview, it is a good idea to start on a particular
topic with an open-ended question, and then once the
interviewee has answered, use a probing question to
gather further information. Closed questions should be
used to confirm understanding and obtain agreement.
Generally speaking, a qualitative interview should last no
more than an hour. A quantitative interview should last
no more than 15 or 20 minutes.
Who can use it?
Interviews should be conducted by skilled individuals in
order to avoid interviewer biases, to generate reliable
information and to deduce trustworthy conclusions.
Within human factors, the technique has many important
uses including selection (page 79), appraisal, training needs
analysis (page 55), system design, system usability (page
28), attitudes, job analysis, task analysis (page 47), and
error and incident investigation.
Finding out more
1. Network Rail, DSM Post-SPAD Interview Process –
Interview Technique and Checklist
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Page 156
Keystroke Level Model
Why is it useful?
More and more jobs on the railways involve the use of
computers, and therefore also usually, keyboards and
associated devices such as a mouse. One of the most
prominent examples is the change in the signaller’s job
from one of using levers, telephone and bell codes in
a traditional mechanical box to one in which nearly all
operations are carried out via a computer terminal in
an IECC signalling centre. One can also point to the
common use nowadays of laptop computers by PICOPs.
Train drivers and other on-train staff will also increasingly
become computer users as new systems such as ERTMS
and GSM-R become operational. Where rail computer
users are performing safety critical tasks it is often
important to know how long the various operations they
must carry out will take. The Keystroke Level Model is a
way of doing this.
What does it do?
Who can use it?
Anyone can learn to use this technique, which requires
very little training.
Finding out more
1. Card S.K. Moran T.P. & Newell A. (1983) The
psychology of human computer interaction, Lawrence
Erlbaum Associates, NJ
2. Salvendy G. (1997) Handbook of human factors and
ergonomics, 2nd edition, John Wiley, Canada
3. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
4. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
The Keystroke Level Model (KLM) is an easy way of
estimating times for serial keyboard-based user tasks.
What does it involve?
KLM provides data on the time it takes for a common
range of individual keyboard operations to be carried out
by users with differing levels of expertise. The technique
also specifies a number of rules for applying the time data
to the keyboard operations and a formula for combining
the results into an overall performance time for the
keyboard task.
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Layout Analysis
Why is it useful?
The layout of controls and equipment in traditional
signal boxes and train cabs has been refined over many
years of experience. The shift to totally new types of
controls and equipment largely made possible through
the introduction of computer-based systems has meant
designers must now to rely on careful analysis to decide
the best layout of controls and displays – especially since
relevant operational experience is often lacking. Layout
Analysis is one technique to support the work of the
designer in this area.
What does it do?
Layout Analysis is a technique for arranging the
components of an interface for simple to moderately
complex tasks in a way that is consistent with the
operator’s mental model, so that the controls, displays
and information are positioned where they are expected.
What does it involve?
The technique involves the analyst working with one or
more representative operators through a number of reorganisations of the interface components. In the first reorganisation, the components are arranged into functional
groups eg in an NX panel, all the points switches would
be located on one place, while level crossing CCTV
controls might be placed in another, relative to the CCTV
display. In subsequent re-organisations, these groups are
arranged in terms of their importance, followed by their
sequence, and finally their frequency of access. The resulting
final arrangement is likely to be the one that best fits
Understanding Human Factors/June 08
the mental model of the user involved in the analysis. It
follows that other operators may differ, and that different
results might also arise for operations during abnormal,
degraded, or emergency conditions, or indeed any
conditions that are not considered during the analysis.
Who can use it?
Anyone with access to a representative (group of) expert
users. Little training is required. However, it will usually be
of benefit to check on the interpretation of the results
with a human factors specialist.
Finding out more
1. Easterby R. (1984) Tasks, processes and display design.
In R. Easterby & H. Zwaga (Eds.), Information Design
(pp. 19-36). John Wiley
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Link Analysis
Why is it useful?
Train drivers and signallers have constantly to shift their
attention from one aspect of their work to another. For
example, a driver has to monitor his various instruments
while maintaining continuous awareness of signals and
what is happening on the track. Designers of new systems
must ensure that the components are organised in a
way that enables the driver or signaller to allocate his
attention in the most appropriate and efficient manner at
any moment in time. This is what Link Analysis can help to
accomplish.
What does it do?
Link Analysis is an easy-to-use and systematic technique
for interface design (page 33). It reveals the actual
relationships for the user between components in
operational or prototype systems. This information can
be used to create a user interface that is simpler to learn
and use.
What does it involve?
Links are movements of the human operator’s focus
of attention eg a driver’s gaze that moves between the
train speed gauge, the AWS sunflower display and a
signal on the route ahead. A link analysis systematically
records these movements as an operator performs their
task eg driving with a new cab design. Data capture can
be achieved by a walk-through or observation analysis.
Where rapid or fine attentional shifts are used by the
operator, data capture may require more specialist eye or
head movement recording apparatus.
Page 157
Part 3: Reference
Techniques
The results of the analysis are usually captured in a
diagram or table of the system components, annotated to
show their importance, frequency, function and sequence
of use. This table can then be studied to suggest more
effective workstation layouts.
Who can use it?
Link analysis is primarily aimed at helping system and
training designers to analyse the layout of panel displays
for eg cab driving, control rooms, signal boxes etc. No
special skills are required, although rehearsal in the
technique is beneficial before its use to collect important
data.
Finding out more
1. Drury C.G. (1990) Methods for direct observation of
performance, In Wilson, J. And Corlett, E. N. Evaluation
of Human Work: A practical Ergonomics Methodology,
2nd Edition, Taylor & Francis, 45–68
2. Kirwan B. (1994) A guide to practical human reliability
Assessment, Taylor & Francis
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Page 158
Murphy Diagrams
Why is it useful?
It is common on the railways to have to find out why
something has gone wrong or why someone made an
error. Simply blaming the culprit may be satisfying but
rarely brings about any underlying improvement. What
is needed is a means for going deeper, to identify the
underlying or root cause. This basic cause may be hard to
find, perhaps originating some time before the problem
becomes visible. The Ladbroke Grove accident provides a
clear example. One of the train drivers did not see a red
signal. Was this just carelessness? Was it simply a result of
his inexperience and lack of training? Examination of the
records showed that a number of other drivers had
also passed this signal at danger. Were they all careless
and badly trained? Or was there some problem with this
particular signal and its sighting? Murphy Diagrams are
one of several ways of analysing incidents to get to the
real cause rather than just the immediately obvious one.
What does it do?
Murphy diagrams is used for analysing human error – for
individual operators or teams. It was developed to analyse
past errors, although it can also be used predictively.
What does it involve?
Murphy diagrams get their name from the common
saying that ‘if anything can go wrong, it will’. Starting with
a comprehensive description of the task, the analyst
classifies each task step into one of a number of decision
making categories. For example, the analyst might assign
the signaller’s step of looking at their NX panel as
“Identification of system state” and then calling a driver
on the cab radio for information as “Interpretation of
situation”. The analyst then breaks each task step down
into successes and failures. As in the Fault Trees technique,
each of the failures are analysed further to identify the
sources of error eg ‘wrong train driver called’ or ‘wrong
code used’. In turn, these errors can be further analysed
into problems that could be addressed eg ‘train code data
unavailable’ or ‘lack of training in train code information’.
Who can use it?
Anyone faced with the task of analysing the causes of
incidents. The technique requires little training. However, it
is not practicable for large and complex tasks.
Finding out more
1. Kirwan B. & Ainsworth L.K (1992) A guide to Task
Analysis, Taylor & Francis
2. Kirwan B. (1992a) Human error identification in
human reliability assessment. Part 1: Overview of
approaches. Applied Ergonomics Vol. 23(5), 299 – 318
3. Kirwan B. (1992b) Human error identification
in human reliability assessment. Part 2: detailed
comparison of techniques. App. Erg, 23, 371-381
4. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
5. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Understanding Human Factors/June 08
Part 3: Reference
Techniques
NASA-Task Load Index
Why is it useful?
For some workers on the railways, workload can be a
real problem. A signaller responsible for a busy section of
track with a complicated layout (a number of junctions
and cross-overs) may find they are stretched to their
limit when also required to deal with two or more gangs
of workers wanting to work on the line. Such a person
is suffering from overload. In contrast, a lookout on a
quiet stretch of track may suffer from the problems of
underload. With little to do and nothing much happening
his level of vigilance may drop, so that his attention has
wandered when a train finally appears and he must warn
the gang of workers on the line. The more accurately
one can characterise and measure workload, the more
appropriate measures can be taken to ensure levels of
workload remain within the optimal range. The NASA
Task Load Index is one of the more sophisticated
techniques for measuring workload.
What does it do?
The NASA Task Load Index (TLX) is one of the best
known, best performing and best respected mental
workload assessment techniques.
What does it involve?
The NASA TLX technique involves observing operators
at work – often via an experimental trial – and then
asking them to rate their experience using six workload
scales. For example, signallers could be observed at
their panels during busy and off-peak periods, or during
simulator sessions at one of the Signalling Training
Centres at Watford or Leeds. The six workload scales are
Understanding Human Factors/June 08
concerned with different kinds of workload and include
mental demand, physical demand, time pressure, effort,
performance effectiveness and level of frustration. The six
scales are then combined to produce an overall workload
rating score. The technique can be used in both software
and paper forms.
Who can use it?
Human factors specialists are required – both to design
or confirm the adequacy of the experimental trial and
to administer the technique. The mental workload
measurement methods developed for use by the UK
railway industry have been largely based on NASA-TLX.
Finding out more
1. Hart S.G. & Staveland L.E. (1988) Development of
a multi-dimensional workload rating scale: Results of
empirical and theoretical research. In P. A. Hancock &
N. Meshkati (Eds.), Human Mental Workload. Elsevier,
Amsterdam
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
4. Vidulich, M. A. & Tsang, P. S. (1986) Technique of
subjective workload assessment: A comparison of
SWAT and the NASA bipolar method. Ergonomics, 29
(11), 1385-1398
Observational Analysis
Why is it useful?
Like interviewing, a lot of observation is carried out on the
railways and for a similar variety of purposes. Managers
observe train drivers and signallers to check on their
competence; LOMs monitor voice recordings to see that
proper communication procedures are being followed;
and many more besides. Again as with interviewing, good
observation requires careful planning and preparation.
Who is to be observed? Doing what? Who should be the
observer? What should the observer be looking for? How
will the observational data be captured? And analysed?
Sound guidance is available for all these questions.
What does it do?
Observational analysis is a technique for the systematic
collection of behavioural information generated during
task performance. It is already in wide use across the
railway industry eg through competence assessment
programmes. Observational analysis is useful in many of
the human factors areas covered by this Guide, especially
design, training and staffing. Very often, observational data
constitutes the groundwork for other human factors
analyses, such as task analysis (page 47), error analysis and
communications (page 106) analysis.
What does it involve?
Observation can take place directly through co-location
with the operators or remotely, via one-way glass or
video. Note, however, that the perceived presence of
observers (whether direct or remote) often changes
the performance that would have otherwise taken
place. Successful application of this technique depends
Page 159
Part 3: Reference
Techniques
principally on adequately structuring the observation
(eg via a checklist or other form); the familiarity of the
observers with what is being observed; and awareness of
observer biases (see sections on selection (page 79) and
risk (page 15)). It should also be noted that the technique
often needs considerable amounts of time to analyse
recorded data – up to a day to process an hour’s worth
of video, for example.
Who can use it?
Anyone who can apply a logical, systematic approach.
However, human factors specialists can be of great
benefit in assisting with structuring the observation so
as to guard against observer bias, as well as in helping to
analyse and interpret the results.
Finding out more
1. Baber C. & Stanton N.A. (1996) Observation as a
technique for Usability Evaluations. In P. Jordan et al
(eds.), Usability in Industry, pp 85-94, Taylor & Francis
2. Kirwan B. & Ainsworth L.K. (1992) A Guide to Task
Analysis, Taylor & Francis
3. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
4. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Page 160
Questionnaire
Why is it useful?
Passengers on trains are frequently asked to fill in a
questionnaire, and many companies use questionnaires to
obtain a variety of types of information from their staff. As
well as gathering factual information, questionnaires are
especially useful in finding out opinions and attitudes. Rail
organisations increasingly realise that staff attitudes are an
important element in both safety and performance. Staff
and public attitudes need particularly to be taken into
account when the organisation is seeking to introduce
change. As with interviews and observation, the familiarity
of questionnaires often means they are used without
sufficient forethought. Clarity of objectives, care in
formulating questions, proper definition of the sample
to receive the questionnaire, choice of appropriate
forms of analysis are all factors that must be considered
if the results obtained are to be trustworthy. Several
questionnaires that have been developed for specific
purposes are described in this Part of the Guide.
What does it do?
Questionnaires are a very powerful and cost-effective
technique for measuring the attitudes and feelings
of a large number of people in a very wide range of
application areas.
What does it involve?
Designing an effective questionnaire – eg to support a
staff feedback process – requires a wide ranging set of
skills. It involves being very clear about the purpose of
the questionnaire, being realistic about the scope and
type of information that can be collected; being clear
about the way the information will be analysed – and the
type of conclusions that can be drawn; and designing and
promoting the questionnaire in a way which maximises
the response rate from its target audience. Piloting the
design – sometimes through several versions – is often
an important component of effective design. One of the
most important considerations is that questionnaires
reflect what people say, which is by no means what
they do. For some applications, analysts may be able to
use pre-defined questionnaires such as SUS (System
Usability Scale), QUIS (Questionnaire for User Interface
Satisfaction) and SUMI (Software Usability Measurement
Inventory). It is often best to check out the results
and apparent implications of questionnaire results by
reflecting them back to representatives of the target
groups, eg via focus groups or workshops.
Who can use it?
While anyone can apparently utilise this technique,
effective questionnaire design, followed by fair analysis
and reporting of resulting data requires a range of skills.
People in most organisations’ HR Departments or human
factors specialists will be able to provide good advice.
Finding out more
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Questionnaires for Distributed
Assessment of Team Mutual
Awareness
is concerned with team task awareness; the second is
concerned with team workload awareness; and the third
is concerned with teamwork awareness.The results of
each questionnaire are combined to provide an overall
measure of team mutual awareness.
Why is it useful?
The questionnaires need to be administered immediately
after a team task session - either operational or simulated.
For example, a training manager might wish to assess the
degree of team mutual awareness within a control room
under different operational conditions. The results could
be used to modify training programme content for new
trainees. Alternatively, a systems design manager might use
the technique in the same setting in order to inform a
new control room design.
As computer-based systems become ever more
widespread on the railways, it becomes important not
only to assess their ease of use, but also how satisfactory
these systems are for their users. A system that is easy
to use does not necessarily lead to user satisfaction. For
example, a system that requires its users to complete
many small steps to carry out a given transaction, may
lead to frustration if there is no way for the more
experienced user to make short-cuts. This questionnaire
provides a standard means for assessing user satisfaction.
Who can use it?
What does it do?
Teamwork is a factor that is increasingly recognised as
an essential ingredient of safe work on the railways.
Train drivers, signallers, station staff, track workers and
others all depend on each other to ensure accidents
do not happen. A feature of teamwork that has been
receiving increasing attention is team awareness. Different
individuals working collaboratively together can only
do this effectively and safely if they share a common
appreciation of the overall situation. Without such a
shared awareness, the right actions cannot be chosen,
and the actions of different participants cannot be
coordinated with each other. This questionnaire has been
designed to help team leaders, trainers and others assess
how well their teams perform in terms of building up a
shared awareness of each other and the situation they
are all working within.
What does it do?
The Questionnaires for Distributed Assessment of Team
Mutual Awareness is a technique for eliciting information
from team members about their degree of mutual
awareness during performance of the team task.
What does it involve?
The technique involves administering three pre-defined
questionnaires covering different aspects of the mutual
awareness of team members. The first questionnaire
Understanding Human Factors/June 08
This is an easy technique for almost any analyst to use
with few training requirements.
Finding out more
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Questionnaire for User Interface
Satisfaction
Why is it useful?
The Questionnaire for User Interface Satisfaction (QUIS)
is an inexpensive and easy-to-use technique for assessing
the acceptability of interfaces. The technique compares
well with SUS, which is even simpler – see Usability testing,
page 28.
What does it involve?
Using QUIS involves asking operators to carry out a
computer-based task and then provide ratings on around
30 separate aspects of their satisfaction with the interface.
These aspects include such items as the confusability of
the information on the screen; the difficulty with which
the interface can be operated; and the frequency with
which the computer provides information about what it
is doing. The technique might be very useful, for example,
in assessing the user acceptability of advanced prototypes
for internal railway information systems such as TRUST or
Page 161
Part 3: Reference
Techniques
the RIMINI planning system; or else external systems such
as passenger ticket machines.
Who can use it?
This is an easy technique for almost any analyst to use
with few training requirements.
Finding out more
1. Chin J.P. Diehl V.A. & Norman K.L. (1988) Development
of an instrument Measuring User Satisfaction of the
Human-Computer Interface. CHI’88
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Sequentially Timed Events
Plotting
Why is it useful?
Many hundreds of incidents occur on the railways
each year that require a formal investigation. Many
managers find themselves involved in carrying out these
investigations. Even apparently simple incidents often
turn out to be much more complex than expected on
closer inspection. Large volumes of data need to be
gathered and scrutinised. These typically include interview
reports, statements, voice recordings, forms, logs, and
so on. Organising all these data into a form where the
all-important question ‘why did this incident happen?’ can
be a daunting challenge. This technique is one means of
arranging and presenting a large volume of data in forms
that aid interpretation and diagnosis.
What does it do?
STEP addresses the problem of how to focus on relevant
information within the high volumes of data generated by
accident or incident investigation. STEP develops a visual
representation of the event, allowing the investigator to
identify what happened and why.
prevent the incident occurring again.
The technique lends itself well to both simple and more
complex investigations. If the incident or accident is
complex, it is useful to split the analysis into sections, for
example, design of equipment, planning of the work, and
the work carried out before the incident.
Who can use it?
The STEP technique can be readily used by rail accident
investigators with little training, although it can be quite
time consuming to complete since it may require a
number of iterations to correctly identify the sequence of
events, and the safety problems or failures.
Finding out more
1. Hendrick, K. & Benner, L. (1987) Investigating accidents
with STEP. New York: Marcel Decker
2. Johnson, S. O., Herrera, I. A., Jersin, E., Rosness, R., Vatn, J.,
Veiseth, M., Tungland, M., Bergersen, C. E. B. (2004) The
Track to Safety Culture (SafeCulture) – a toolkit for
operability analysis of cross border rail traffic, focusing
on safety culture. SINTEF for the UIC
What does it involve?
Using STEP involves creating a diagram of the main
actors, events and relationships plotted against time.
The resulting matrix can answer ‘where’, ‘what’, ‘when’
and ‘how’ questions involved in an accident or incident
investigation. The information from the analysis of
safety problems can be used to identify appropriate
recommendations and actions, which if carried out would
Page 162
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Situation Awareness for SHAPE
Why is it useful?
Compared to earlier times, today’s railway worker has
to keep a lot more in mind at any one time. Think, for
example, of signallers in a modern power box. The
length of the section for which they are responsible may
be many times longer than that of their predecessors
working in a mechanical box. They are also likely to be
monitoring the movements of many more trains. Panel
or IECC technology helps them to do this, but they
must also build up and maintain a mental picture of
the situation. This mental picture or model is essential
to be able to project how the situation will develop
into the future, thus enabling timely action to ensure a
smooth service without unnecessary disruption or other
problems. Train drivers also need to maintain awareness
of complex and rapidly changing situations, and the
demands on them will increase as new communication
and information systems provide them with much more
information than they have had historically.
Designers, trainers, managers and others have an
increasing need to assess the demands for situation
awareness of new or changed jobs. The two techniques
described here were developed in the field of air
traffic management to help analyse and assess situation
awareness aspects of tasks.
What does it do?
The Situation Awareness for SHAPE (SASHA)
methodology consists of two complementary techniques
for assessing an operator’s situational awareness (SA)
Understanding Human Factors/June 08
(page 33). The methodology was developed for air traffic
control but could be developed for use with real-time
display-based railway tasks quite well, eg signalling, control
room operation, cab driving.
What does it involve?
The first of the two SASHA techniques is SASHA_L.
This generates probes to the operator during the task,
which can be either operational or simulated. These
probes are issued either by the analyst or (if the task is
simulated) possibly by computer. In either case, there is
no requirement for the task to be stopped. The idea is
to measures both accuracy and response times to the
probes to help infer the operator’s situation awareness.
For example, SASHA_L probes to a signaller might
include questions such as: “Which train needs to be
attended to next?”; “Which train is moving faster?”; or
“Which train would benefit from a direct route?”.
Finding out more
1. Jeannot, E. Kelly, C. & Thompson, D. (2003) The
development of Situation Awareness measures in ATM
systems. EATMP report. HRS/HSP-005-REP-01
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
The second technique is SASHA_Q. This is a short
questionnaire that is presented to the operator
immediately after task performance. The questions are
answered using 5-point scales from ‘Never’ to ‘Always’ or
‘Often’. For example, SASHA_Q questions to a signaller
might include: “Did you have the feeling that you were able
to plan and organise your work as you wanted?”; and “Were
you surprised by a call that you were not expecting?”
Who can use it?
While the SASHA techniques are simple to apply, the
training time could be high for the non-computerised
version. This reflects the time taken for the analyst (who
should be a subject matter expert) to become proficient
at generating relevant SA probes during the task.
Page 163
Part 3: Reference
Techniques
Situation Awareness Rating
Technique
to the operator; and the ability of the operator to
understand the situation.
Why is it useful?
SART is quick and easy to use with few training
requirements.
Most safety critical jobs on the railways require the
person doing the job to maintain a high level of
awareness of the situation that not only provides a
picture of the present situation but also how the situation
will develop in the future. Where two or more people
need to collaborate on a task in a dynamic situation,
they can only do this as long as they share a common
awareness of the aspects that affect them both. This
technique is designed to help designers and others assess
the shared situation awareness requirements of a task.
What does it do?
The Situation Awareness Rating Technique (SART)
is a technique for assessing team members’ situation
awareness (page 33). It can provide useful input to team
workstation design and training.
What does it involve?
SART involves the analyst eliciting ratings on ten aspects
of situational awareness from team members immediately
after they have completed a task. The ten aspects include
factors such as the perceived familiarity, complexity and
instability of the situation, and quantity and quality of
information.
There is also a quicker version of the technique which
collapses the ten SART factors into just three. These
three groupings are concerned with demands on the
operator’s attention; the attentional resources available
Page 164
Who can use it?
Finding out more
1. Selcon S.J. & Taylor R.M. (1989) Evaluation of the
Situation Awareness Rating Technique (SART) as a
tool for aircrew system design, Proceedings of AGARD
Symposium on Situational Awareness in Aerospace
operation, Copenhagen
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
4. Taylor R.M. (1990) Situational Awareness Rating
Technique (SART): The development of a tool for
aircrew systems design. In Situational Awareness in
Aerospace Operations (AGARD-CP-478) pp3/1
–3/17, Neuilly Sur Seine, France: NATO-AGARD
Soft Systems Methodology
Why is it useful?
Recent history in the railway industry (and in many
other industries) shows that it is exceedingly difficult to
design and introduce complex systems. Of course, under
the impact of new technologies, systems have become
much more complex in recent years and are destined to
become still more complex. The most common cause of
failure in developing complex systems lies in the definition
of the system requirements. People across an organisation
often do not agree what the system is for or how it
should work. Often, they are never all consulted. The
Soft Systems Methodology is a well-worked out process
by which stakeholders in a new system work together
to build up an agreed conceptual model of the system:
what it should be and what it should do. This conceptual
model then provides a more reliable foundation for the
subsequent detailed specification of the system.
What does it do?
Soft Systems Methodology (SSM) is an approach to
general problem solving that emphasises interpretation
and meaning rather than hard measurement and
quantitative analysis. It is based on systems thinking and is
especially useful for user requirements analysis, but it has
also been used to support organisational restructuring
and performance indicator development.
What does it involve?
The SSM analyst starts with a stakeholder analysis and
then creates conceptual models of system activities
based on root definitions of key system elements. A
cornerstone of SSM is its emphasis on taking a number
Understanding Human Factors/June 08
Part 3: Reference
Techniques
of different viewpoints of the system. The stakeholders
provide two such viewpoints – users and beneficiaries.
Other viewpoints include the system function, its context
and its environmental constraints. When the models
have been developed, they are compared with reality
in order to determine how any gaps between the two
can be reduced. SSM underlines the view that systems
arises out of an organisational need and must fit into
an organisation and its culture by servicing multiple
perspectives. In the railways, the technique would be
suited to the development of a wide range of processes
for which user requirements needed to be defined in a
multi-stakeholder context eg the Rule Book or possession
management.
Who can use it?
SSM analysts do not need to be trained in any particular
discipline in order to use the technique, although an
affinity for systems thinking helps a lot. However, the
technique itself requires significant training time and
a complex application may take several months to
complete.
Finding out more
1. Checkland P. (1989) Soft systems methodology, in
Rosenhead J. (Ed.), Rational analysis for a problematic
world, 71-120, John Wiley
2. Checkland P. & Scholes J. (1990) Soft systems
methodology in action, John Wiley
3. Checkland P. (1981, 1998) Systems Thinking, Systems
Practice, John Wiley
Software Usability Measurement
Inventory
Why is it useful?
Increasingly, railway workers are users of software
systems, even though they may not always be aware of
software embedded in a piece of equipment that is not
obviously a computer. In very many situations safety may
depend on the usability of the software. This technique
is designed to assess the usability of software measured
against a set of attitude statements.
This technique would be very useful as part of the testing
suite of anyone responsible for the design, usability or
acceptability of any new software system proposed for
deployment within the railways.
Who can use it?
What does it do?
The Software Usability Measurement Inventory (SUMI)
is a commercial, comprehensive and quick usability
testing (page 28) technique that provides a much more
fine-grained analysis than its even quicker (and free)
stablemate, SUS (System Usability Scale).
What does it involve?
SUMI involves people using a software application and
then rating their experience in terms of 50 attitude
statements. Each statements requires one of three
responses: ‘agree’, ‘don’t know’ or ‘disagree’. Examples of
SUMI statements are:
• This software responds slowly to inputs.
• The instructions and prompts are helpful.
• The way that system information is presented is clear and
understandable.
• I would not like to use this software every day.
Understanding Human Factors/June 08
SUMI’s output includes a global usability score, several
more specialised usability scores for aspects such as
the software’s helpfulness, control and learnability, and a
benchmark score that allows the analyst to know if any
aspects of the software under test are very different from
expected norms.
Anyone - requires very little training. SUMI is a
commercial product that requires a licence fee of over
€1,000.
Finding out more
1. Kirakowski J. (1996) The Software usability
measurement inventory: background and usage. In P.
Jordan, B. Thomas, & B. Weerdmeester (eds) Usability
Evaluation in Industry. Taylor & Francis
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
4. SUMI website (as of May 2008)
www.ucc.ie/hfrg/questionnaires/sumi/index.html
Page 165
Part 3: Reference
Techniques
Subjective Workload Assessment
Technique
Why is it useful?
Safety can be affected by workload that is too low as well
as too high. Workload is essentially the experience of the
person doing the work. What might be high workload
for one person may be perfectly tolerable to another.
This workload assessment technique relies upon asking
workers to assess their own workload using rating scales
that they are first helped to develop specially for their
specific task.
What does it do?
Subjective Workload Assessment Technique (SWAT)
is a commonly used alternative to NASA TLX, offering
greater insight into the reasons for high workload (page
125). However, it is not very sensitive to mental workload.
What does it involve?
SWAT involves measuring three aspects of operator
workload: time load is concerned with time limits and
the extent to which tasks must be carried out together;
mental effort load is concerned with attention and
mental processing; and stress load is concerned with
operator characteristics that are known to affect task
performance, (eg training, fatigue).
The process by which these three measures are taken
lends considerable power to the results, but is a little
involved. In particular, the scale to be used must be
customised for each task of interest by the operators,
who then perform the task and use the scale to rate the
Page 166
workload they experience.
Who can use it?
The technique is time-consuming to set up, and
preferably requires an analyst with previous experience
with it. A human factors specialist is of benefit to assist
with technique administration and interpretation.
Finding out more
1. Cha D.W. (2001) Comparative study of subjective
workload assessment techniques for the evaluation
of ITS-orientated human-machine interface systems.
Journal of Korean Society of Transportation. Vol 19 (3),
45-58
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
7. Vidulich M.A. & Tsang P.S. (1986) Collecting NASA
Workload Ratings. Moffett Field, CA. NASA Ames
Research Center
8. Vidulich M.A. & Tsang P.S. (1986) Technique of
subjective workload assessment: A comparison of
SWAT and the NASA bipolar method. Ergonomics, 29
(11), 1385-1398
2. Dean T.F. (1997) Directory of Design support
methods, Defence Technical Information Centre,
DTIC-AM. MATRIS Office, ADA 328 375, September
3. Hart S.G. & Staveland L.E. (1988) Development of
a multi-dimensional workload rating scale: Results of
empirical and theoretical research. In P. A. Hancock &
N. Meshkati (Eds.), Human Mental Workload, Elsevier,
Amsterdam
4. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
5. Reid G.B. & Nygren T.E. (1988) The subjective
workload assessment technique: A scaling procedure
for measuring mental workload. In P. S. Hancock & N.
Meshkati (Eds.), Human Mental Workload, Elsevier,
Amsterdam
6. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
Understanding Human Factors/June 08
Part 3: Reference
Techniques
System Usability Scale
Finding out more
Why is it useful?
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
System usability is central to safety. This is true whether
the system in question is a trackman with the right tools
for changing a fishplate, the operator and their road-rail
machine, a signaller with an IECC workstation, a driver of
a Voyager train, a train crew operating the refreshments
trolley, or passengers trying to find their way in and out
of Pendolino toilets. Techniques for assessing usability vary
greatly in the detail they provide and the effort required
for their use. This is perhaps the simplest usability
assessment technique currently available.
What does it do?
The System Usability Scale (SUS) is a simple
questionnaire, used to rate the usability (page 28) of an
existing system or advanced prototype product.
What does it involve?
SUS consists of ten usability statements with which
participants are asked to indicate the extent of their
agreement after completing a task using the system under
test. Answers are coded according to a simple procedure
and a total score is calculated for overall usability.
Who can use it?
Anyone - requires very little training and very little time
to apply. It is the simplest and quickest means of assessing
the overall usability of a product or device.
Understanding Human Factors/June 08
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Systematic Human Error
Reduction and Prediction
Approach
Why is it useful?
People working on the railways perform many different
types of task. As with all human acts, errors can occur. In
many situations it is desirable to know in what ways the
task can go wrong, how probable it is it will go wrong,
and how much it matters if it goes wrong. For example,
many human errors on the railway are related to
communication. Communications can go wrong in many
ways: you can forget to tell somebody something they
need to know, you can mix up numbers, you can speak
unclearly, the person you are talking to may interpret
what you say in a way other than what you intended, and
so on. If you are talking to a colleague about next week’s
roster, little harm may result from the error. If you are
an IWA reporting your position to a signaller, then any
misunderstanding might put you in serious danger. This
technique allows potential errors to be identified and
described in a way that allows measures to be taken to
reduce the risk of this error actually occurring.
What does it do?
Systematic Human Error Reduction and Prediction
(SHERPA) is one of the best techniques that has been
produced for human error analysis, prediction and
reduction.
Page 167
Part 3: Reference
Techniques
What does it involve?
SHERPA starts with a task analysis (page 47) and,
commencing with the lowest level, classifies operations
into one of several types of human activity, including
action, retrieval, checking, selection or communication.
For each classification made, the analyst decides the likely
errors, their potential consequences, the recovery path,
and the probability and criticality of their occurrence.
Finally, the analyst considers how the identified errors
could be best avoided eg by equipment re-design, training,
new procedures, or organisational changes.
Who can use it?
Human factors specialists working in conjunction with
subject matter experts.
Finding out more
1. Bass A. Aspinal J. Walter G. & Stanton N.A. (1995) A
software toolkit for hierarchical task analysis. Applied
Ergonomics. 26 (2) 147-151
2. Embrey D.E. (1986) SHERPA: A systematic human
error reduction and prediction approach. Paper
presented at the International Meeting on Advances in
Nuclear Power Systems, Knoxville, Tennessee
3. Embrey D.E. (1993) Quantitative and qualitative
prediction of human error in safety assessments.
Institute of Chemical Engineers Symposium Series, 130,
329-350
4. Hollnagel E. (1993) Human Reliability Analysis: context
and control, Academic Press
Page 168
5. Kirwan B. (1990) Human reliability assessment. In J. R.
Wilson & E. N. Corlett (eds.), Evaluation of human
work: a practical ergonomics methodology (2nd ed. pp.
921-968), Taylor & Francis
6. Kirwan B. (1992) Human error identification in human
reliability assessment. Part 2: detailed comparison of
techniques. Applied Ergonomics, 23, 371-381
7. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
8. Stanton N.A. (1995) Analysing worker activity: a
new approach to risk assessment? Health and Safety
Bulletin, 240, (December), 9-11
9. Stanton N.A. (2002) Human error identification in
human computer interaction. In: J. Jacko and A. Sears
(eds) The Human-Computer Interaction Handbook.
Lawrence Erlbaum Associates, NJ
10.Stanton N.A. & Baber C. (2002) Error by design:
methods to predict device usability. Design Studies, 23
(4), 363-384
11.Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
12.Stanton N.A. & Young M. (1998) Is utility in the mind
of the beholder? A review of ergonomics methods.
Applied Ergonomics. 29 (1) 41-54
Task and Training Requirements
Methodology
Why is it useful?
One of the main means of helping to ensure that railway
staff can do their jobs reliably is training. Tasks vary in
terms of what type, and how much, training they require.
There are many forms of training, and finding the right
form to develop the knowledge and skills required needs
careful analysis. Once learned, tasks also vary in whether
they need to be re-learned at intervals. Proverbially, once
you have learned to ride a bike you never forget how. But
other tasks, for example, those requiring the application
of complex rules, can get ‘rusty’ if not regularly practiced.
This methodology is one of a number designed to analyse
training requirements, but has particular features that
make it valuable in the identification of training solutions
in a team context.
What does it do?
The Task and Training Requirements Methodology
(TTRAM) is a kind of training needs analysis (TNA) (page
55) and is oriented towards TNA for teams. It is very
useful for identifying tasks prone to skill fade (page 64)
as well as training gaps. It identifies the underlying skills
associated with each task and helps with the identification
of training solutions. Finally, it rates the level of teamwork
required for each task step.
What does it involve?
TTRAM requires an analyst to apply two pre-defined
scales for skill fade and skill practice to task information
gained from interviews with subject matter experts.
Understanding Human Factors/June 08
Part 3: Reference
Techniques
The skill fade score addresses task difficulty, degree of
prior learning and frequency of task performance. The
practice score addresses the amount, frequency and
quality of practice. The results are two scores which are
compared to identify training gaps. For example, the
technique could help to determine refresher intervals for
different signalling panels in a signal box, or for different
maintenance tasks for train fitters in a depot. TTRAM
provides further tools to help identify how training gaps
can be filled with suitable training technologies and media.
Who can use it?
TTRAM requires considerable access to subject matter
experts and greatly benefits from human factors or
training specialists. It is also very time-consuming to use.
Finding out more
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
3. Swezey R.W. Owens J.M. Bergondy M.L. & Salas
E. (2000) Task and training requirements analysis
methodology (TTRAM): An analytic methodology for
identifying potential training uses of simulator networks
in teamwork-intensive task environments. In J. Annett
& N. Stanton (eds) Task Analysis, pp150 – 169, Taylor &
Francis
Understanding Human Factors/June 08
Task-Centred System Design
Why is it useful?
As work systems in trains, signalling, stations, and
track work become ever more complex, it becomes
increasingly difficult to ensure that the system will
do what it is supposed to do, safely and reliably. This
methodology is more than a technique, but is a whole
design approach that enables a thorough assessment to
be made of an emerging system design.
What does it do?
Task-Centred System Design (TCSD) is a methodology
for evaluating new system design that incorporates the
principles of user-centred design (page 25) from the start.
What does it involve?
TCSD involves a multi-disciplinary design team going
through a number of analytic phases.The team first uses
interviews and workshops to identify user groups and
example tasks for the proposed new system. Next the
team develops a number of representative task scenarios.
One by one, these are used to test the design until the
team is satisfied. The team collectively walks-through the
scenarios, using role-play and imagination to ‘experience’
the system from the point of view of the identified users.
The output of the technique is a refined system design
which can also contribute to job design as well as training
requirements specification.
Who can use it?
The power of the technique depends on the quality
of the design team members and their ability to select
and utilise the right scenarios to test the evolving
design. It also depends on their imagination when they
walk-through the scenarios. For the most accurate and
reliable conclusions, the team needs to include subject
matter experts, designers, human factors specialists and
operators.
Finding out more
1. Greenberg S. (2003) Working through Task-Centred
System Design. In D. Diaper & N. Stanton (Eds) The
Handbook of Task Analysis for Human Computer
Interaction. Lawrence Erlbaum Associates
2. Lewis C. & Reiman J. (1993) Task centred user interface
design: A practical introduction. Boulder, CO: University
of Colorado. Shareware book available from ftp.
cs.colorado.edu/pub/cs/distribs/clewis/HCI-DesignBook
3. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
4. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
One application of TCSD on the railways might be in the
design of a new control room that can cope with normal,
abnormal, degraded and emergency conditions. Another
example is the design of a new multi-section signal box.
Page 169
Part 3: Reference
Techniques
Team Cognitive Task Analysis
Why is it useful?
Two trends are noticeable on the railways. One is that
work is becoming more cognitive (ie mental) rather
than physical. The other is that the numbers of groups
of railway workers that can be considered as teams is
increasing as new technologies let different occupations
work together in a more direct and responsive manner.
For example, a train driver and a signaller can already be
considered as a team while the driver’s train is passing
through the signaller’s section. As new technologies
such as ERTMS enter service, this working relationship
will become even closer and more dynamic. The form
of analysis described here integrates several techniques
to enable the tasks carried out by teams rather than
individuals to be examined to support design and training
initiatives.
What does it do?
Team Cognitive Task Analysis (TCTA) adapts and
combines cognitive task analysis (page 49), Critical Decision
Method (CDM) and Team Decision Requirements Exercise
(TDRE) to improve understanding of how a team makes
decisions as they work together on a team task. It is
particularly useful in understanding teamworking (page
107) and in helping to design team workflows and
diagnose team training needs.
team task and interviewing team members. Observed
incidents and decisions made by team members are
classified in terms of several factors that underpin team
task performance such as shared situation awareness and
planning skills. For each incident classified in this way, the
analyst documents the information and resources used by
the team member(s) in making their decisions. The analyst
also records any difficulties experienced by the decision
makers eg failures in communication or technology. The
output of the technique is a decision requirements table
which sets out how more effective decisions can be made
in similar team tasks and environments in the future.
TCTA could be profitably employed in the many areas
of the railway industry that rely on teamwork, eg railway
control rooms, signal boxes, trackwork. The outcome
would be improvements to the training of teams, and/
or changes to the team structure or procedures that
support the teamwork.
Who can use it?
TCTA requires analysts who are well trained and
experienced in the technique and its related techniques
(CDM and TDRE). Such people are likely to be human
factors specialists.
Finding out more
1. Klein G. (2000) Cognitive Task Analysis of Teams.
In J. M. Schraagen, S. F. Chipman, V. L. Shalin (Eds)
Cognitive Task Analysis. pp417-431, Lawrence Erlbaum
Associates
2. Klein G. & Armstrong, A.A. (In Press) Critical Decision
Method. In Stanton et al (Eds) Handbook of Human
Factors and Ergonomics methods, Taylor & Francis
3. Klein G.A. Calderwood R. & MacGregor D. (1989)
Critical Decision Method for Eliciting Knowledge. IEEE
Transactions on Systems, Man and Cybernetics, 19(3),
462-472
4. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
5. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
What does it involve?
TCTA should be carried out for some defined purpose,
such as error reduction, improving performance efficiency
or making changes to team numbers or quality. A TCTA
analysis involves observing performance of an existing
Page 170
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Team Decision Requirements
Exercise
Why is it useful?
In some important situations, especially such as after
a major incident, decisions are not made not just by
individuals but by groups of people working together as
a team. In preparing for such instances it is important
to have worked out how the team should operate
together to make the decision, what information they
will require, how and when they should communicate,
and so on. This technique is one way of gaining a
necessary understanding of these team decision making
requirements.
What does it do?
The Team Decision Requirements Exercise (TDRE) is
a relatively powerful, easy-to-use structured interview
technique for revealing how critical decisions are made
by a team during task performance. It is useful for
understanding task requirements for both system and
training design. TDRE is a variation of Critical Decision
Method (CDM) (itself a development of Critical Incidents
Technique CIT). The technique is also a component of
Team Cognitive Task Analysis.
What does it involve?
TDRE involves carrying out one or more group
interviews with the team under analysis. The team is
probed with questions aimed at eliciting their approach
to decision making and its associated information sources,
difficulties, errors and potential improvements. The output
of the technique is a decision requirements table which
Understanding Human Factors/June 08
sets out how more effective decisions can be made in
similar team tasks and environments in the future.
As for Team Cognitive Task Analysis, TDRE could be
profitably employed in the many areas of the railway
industry that rely on teamwork, eg railway control
rooms, signal boxes, trackwork. The outcome would
be improvements to the training of teams, based on a
better understanding of the objectives, roles, procedures,
difficulties, communications and expertise of existing
team members.
Who can use it?
TDRE requires two analysts who are skilled in the
technique – and particularly interviewing – in order to
get the most out of it. Other than prior experience with
the technique, it is of some benefit if the analysts are also
subject matter experts.
Finding out more
1. Klinger D.W. & Hahn B.B. (In Press) Team Decision
Requirement Exercise: Making Team Decision
Requirements Explicit
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Team Workload Assessment
Why is it useful?
The workload of an individual worker on the railways
is obviously important in planning jobs, systems, and so
on. One reason why teamwork can be so effective is
that workload can be shared amongst members of the
team, helping to ensure that no one person is either
over-loaded or under-loaded. In setting up working
arrangements and system support to teams it is valuable
to be able to estimate the workload on the entire team
under varying working conditions. This technique helps
designers and planners estimate team workload, as well
as individual workload.
What does it do?
Team Workload Assessment (TWA) is a technique that
extends the respected NASA-TLX workload assessment
instrument for use in team settings.
What does it involve?
In addition to the standard NASA-TLX procedure, the
TWA extension requires team members to make
two assessments: one is of their own workload while
the second is of the workload of the whole team. The
assessments from each are team member are then
combined, using a prescribed formula, to create a team
task workload score.
The mental workload measurement methods developed
for use by the UK railway industry have been largely
based on NASA-TLX. The TWA is capable of providing
a highly relevant extension, increasing its utility for
railway environments – not only control rooms and
Page 171
Part 3: Reference
Techniques
signal boxes where team members are co-located, but
also for distributed teams such as drivers, signallers
and trackworkers who must work together eg for the
duration of an incident.
Who can use it?
Human factors specialists are required – to design or
confirm the adequacy of the experimental trial and to
administer the technique.
Finding out more
1. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
2. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Teamworking Improvement
Process
Why is it useful?
Many safety critical tasks on the railways depend on
good quality teamwork. For example, the COSS, lookout
and site warden must work closely together to ensure
the safety of a gang of track workers. Good teamwork
depends on many factors: training, motivation, willingness
to communicate, openness to learning, and so on. There
is almost always scope for improving teamwork. This
methodology takes a company in the railway industry
through assessment and diagnosis activities, which lay the
foundations for choosing the most cost-effective method
for improving teamwork.
Who can use it?
The technique can be implemented by a small team
of railway line managers, a training or HR manager
and front-line staff. While the technique requires little
resource or training to implement, the teamworking
interventions generated by the process may require
significant organisational resource to implement should
the organisation decide to do so.
Finding out more
1. Gregory D. & Shanahan P. (2004) Teamworking best
practice in the railway industry: The Journey Guide,
Gregory Harland Ltd, for RSSB, Euston
What does it do?
The Teamworking Improvement Process (TIP) was
developed specifically for the railway industry from an
analysis of best practice in teamworking (page 103). It is a
highly practical process that requires a small project team
drawn from the organisation under analysis.
What does it involve?
TIP involves a three phase process that takes place
over 12 months or so. The technique provides survey,
spreadsheet and workshop tools to perform a diagnosis
of the organisation’s teamworking proficiency, to analyse
the best improvement route, and then to measure the
benefits following implementation of the teamworking
interventions selected.
Page 172
Understanding Human Factors/June 08
Part 3: Reference
Techniques
User Trial
Why is it useful?
Any change or innovation on the railways always entails
an element of risk. The railways comprise a complex
system of people, rules, infrastructure, technology, rolling
stock, procedures and working practices. How well
any change will work is often hard to predict. Changes
often have unexpected and unwelcome side effects.
The introduction of any new or changed feature of
work which is safety critical must be properly evaluated
through a carefully controlled user trial.
What does it do?
A user trial is structured session in which a new system,
process or procedure is tested with its intended users for
the purpose of generating feedback to its designers. Types
of feedback may include usability (page 28), workload
(page 125) and situation awareness (page 33).
may be needed to identify some of the factors for study
and/or control for human biases within the user trial,
specialists may be needed to administer the selected
technique(s), and designers will be needed to configure
the trial system and consider the trial’s conclusions.
Finding out more
1. Salvendy G. (1997) Handbook of human factors and
ergonomics, 2nd edition, John Wiley, Canada
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Walk-through Analysis
Why is it useful?
Developing new systems for the railways requires major
investment in time, effort and money. The later the need
for any change to a developing system, the greater the
cost entailed in going back and doing things differently.
The earlier any potential problems with, say, usability or
maintainability, are identified, the better. This technique has
been created to assist designers and developers pick up
any problems at the earliest opportunity when the costs
of remedial action are lowest.
What does it do?
Walk-through Analysis is a technique in which qualified
personnel ‘walk-through’ a representative task using a
prototype system or process. It is a powerful, cheap and
re-usable technique for evaluating a new design. It is
limited only by the imagination and creativity of its users.
What does it involve?
What does it involve?
Conducting an effective user trial involves setting clear
trial objectives, defining representative trial tasks, choosing
representative users, selecting and applying appropriate
within-trial techniques (eg SUMI, ISA, SASHA), interviewing
participants (if necessary), analysing and summarising the
trial data, and making clear, practicable recommendations.
A walk-through involves someone stepping through a
task, facilitated by a system or process, explaining their
reactions, decisions and actions as they go. The technique
is useful to explain practice with systems or to examine
likely practice in future ones. Importantly, the technique
can be used effectively (subject to the participants’
imagination) even with no actual task or system present,
which makes it very useful at very early stages of design.
Who can use it?
Designing and running a user trial requires a wide range
of skills. Statisticians may be needed to design balanced
trials that can simultaneously examine several different
factors of interest. Subject matter experts may be needed
to design representative tasks. Human factors specialists
Understanding Human Factors/June 08
Page 173
Part 3: Reference
Techniques
Who can use it?
Design teams, including subject matter experts and/
or experienced operators of existing or prospective
replacement systems.
Finding out more
1. Kirwan B. & Ainsworth L.K. (Eds) (1992) A guide to
Task Analysis, Taylor & Francis
2. MOD (2006), MAP-01-011 Human Factors Integration
Technical Guide (Annex 3), Sea Systems Group, TESSSG-ShipDes, Defence Procurement Agency, Bristol
3. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Why-Because Analysis
Why is it useful?
Investigators of incidents on the railways are required to
follow laid down procedures. In particular, investigators
are required to go beneath the obvious causes and
attempt to unearth the root cause – the fundamental
reason why something went wrong. For example, as soon
as it becomes apparent that a railway worker made an
error, investigations in the past were happy to stop at that
point, labelling the cause of the incident as ‘human error’.
To simply label a cause as human error gives no indication
as to what measures are needed to prevent the error
(and the resulting incident) from happening again. The
conclusion ‘human error’ should be the beginning of
an investigative process, not the end. The interesting
question is ‘why did this person make this error at this
time?’ Further digging may reveal deficiencies in, say, the
training process, the process for assessing competency, or
source of stress in the worker’s private life. This technique
is a simple means of going beyond the obvious to find a
cause which can be the subject of remedial action.
What does it do?
Why-Because Analysis is an easy-to-use form of root
cause analysis that was developed specifically for line
managers in the railway industry to investigate past near
misses. The technique also allows the development of
countermeasures, and tracks their implementation status.
What does it involve?
Why-Because Analysis involves creating a diagram that
represents the factors involved in a past incident eg
a near miss. The diagrams are a little like Fault Trees.
Page 174
For each problem (eg trackworker’s late escape from
the path of a train), they capture the direct causes (eg
lookout at wrong sighting distance + train horn masked
lookout’s warning ), the external causes (eg trackwork
task running late + insufficient trackwork resource to do
the job); and countermeasures – ie those interventions
which would, if made, prevent recurrence of the incident
(eg implementation of prohibited red zone).
For complex incidents where a simple Why-Because
diagram is not enough, a simple documentation scheme
can be used. This scheme makes extensive use of the
hyperlink facility in MS Office™ tools to connect the
elements of the diagram with supporting reports and
recommendations. This also makes it very easy to track
progress against any recommended countermeasures
which have been accepted for action.
Who can use it?
This technique was developed specifically for line
managers in the railway industry and is easy to
understand and use. However, the technique requires
reasonable proficiency with MS Office™ tools.
Finding out more
1. Braband J. & Brehmke B. (2002) Human factors
application area of Why-Because Graphs to Railway
Near-Misses, In Workshop on the Investigation and
Reporting of Accidents 2002, (IRIA 2002) Ed C. W.
Johnson. GIST Technical Report F2002-2, Dept Of
Computing Science, University Of Glasgow, Scotland
2. Ladkin P. (2001) Causal System Analysis – Formal
Reasoning About Safety and Failure, Uni. of Bielefeld
Understanding Human Factors/June 08
Part 3: Reference
Techniques
Workshops
Why is it useful?
In many areas of change and development in the railways
the views of a wide range of stakeholders need to be
taken into account. Meetings are the usual means for
bringing together the various interested parties. But
meetings tend to be very time-consuming and can
sometimes lack focus. Often meetings finish without the
desired objectives having been reached. The workshop
can be considered as a much more structured meeting.
It should have been carefully planned, its objectives
defined properly and the steps required to achieve these
objectives designed in detail. The required outputs from
the workshop will have been specified and a trained
facilitator will take responsibility for the conduct of the
workshop.
What does it do?
A workshop is a structured discussion that can be
used for a very wide range of applications. When it
is composed of the right stakeholders and run by
experienced facilitators, a workshop is one of the
most powerful tools available for eliciting, refining and
prioritising information. The technique is also invaluable
for securing agreement and building trust between
project stakeholders.
What does it involve?
A workshop involves the design and execution of
a process by which a group of stakeholders share
information about a set of topics. The process should
always have a clear aim and a series of logical stages
through which the facilitators move the participants.
Understanding Human Factors/June 08
Workshops tend to take place of longer periods of time
than Focus Groups and can run from a half-day through
to several days duration. They may also involve more
participants – typically 8 to 20 or so.
The longer durations and more people mean that
workshops can profitably use combinations of plenary
groups (everyone) and syndicates (sub-groups) for
different kinds of activity. Often, syndicates are used to
work on different aspects of a complex problem. At
other times, they can be used to work on the same
problem so that the group as a whole ends up with more
options. Either way, syndicate work is usually shared and
discussed in the plenary group, which can also be used
for brainstorming and keynote presentations.
Who can use it?
Anyone can use this technique. However, workshops
benefit greatly from careful design so that a clear
process is put in place that will lead to a clearly defined
objective. This helps identify the exact requirements for
the participants. Experienced facilitators make a huge
difference to the value, enjoyment and ultimate success of
workshops.
Finding out more
1. Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods: A
Practical Guide for Engineering and Design. Ashgate,
Aldershot. ISBN 0-7546-4660-2 (HBk), 0-7546-4661-0
(PBk)
Page 175
Part 3: Reference
Techniques
Page 176
Understanding Human Factors/June 08
understanding
Jargon buster
human factors
understanding
Jargon buster
human factors
Part 3: Reference
Jargon buster
Jargon buster
Ability
‘Ability’ is used loosely both in the literature and everyday
life either to refer to actual performance, or else (more
commonly) to a capability for performance. In the former
case, ‘ability’ can be replaced with ‘skill’; in the latter case,
‘ability’ can be replaced with ‘aptitude’.
Abnormal operating conditions
State of continuing railway operations with specific planned
changes to its configuration or equipment (such as special
events, engineering works in traffic hours or within station
public areas).
Accident
An unintended event, often resulting in injury and/or loss.
Accidents usually have several antecedents, many of which
may have originated some considerable time before the
accident, and whose combination was never anticipated.
ACTA
Applied Cognitive Task Analysis - see Techniques.
Anthropometry
The measurement of human body dimensions.
Aptitude
The natural propensity of an individual to acquire and utilise
specific knowledge, skills and attitudes.
Arousal
A state of body and mind that sensitises an individual to the
search for, and receipt and processing of, information, and
which prepares an individual for action. Too much arousal
can result in stress, which can overload the individual and
Understanding Human Factors/June 08
render them increasingly unable to deal with, or act on,
information.
CIT
Attitude
A manner of acting, conduct or interactional style which is
representative of feelings, opinions or beliefs.
Cognitive mapping
Cognitive mapping – see Techniques.
Behaviour modification
The replacement of undesirable behaviour with desirable
responses using programmes that redirect existing behaviour
via various kinds of conditioning. The conditioning that works
best is based on reward for desired responses, rather than
punishment for undesired responses. Behaviour modification
focuses only on behaviour rather than the thinking that lies
behind the behaviour. The success of behaviour modification
programmes can be considerably enhanced by taking the
thinking side into account as well. In a clinical setting, this
multiple approach is called cognitive behaviour therapy.
Brainstorming
Brainstorming – see Techniques.
Brightness
The subjective response to luminance in the field of view,
dependent upon the adaptation of the eye.
CDM
Critical Decision Method - see Techniques.
Checklist
Checklist – see Techniques.
CIE
Commission International de l’Eclairage, (International
Commission on Illumination). An international organisation
responsible for colour and light measurement standards.
Critical Incident Technique - see Techniques.
Cognitive task analysis
The elicitation and representation of the conscious thought
processes that underlie the performance of tasks and
thinking skills needed to respond to complex situations. (See
Applied Cognitive Task Analysis in Techniques).
Cognitive walk-through
Cognitive walk-through – see Techniques.
Contrast
Subjectively used, contrast describes the perceived difference
in appearance between two parts of a visual field seen
simultaneously or successively. The difference may be one
of brightness or colour or both. Objectively used, the term
expresses the measurable difference in luminance between
two parts of the visual field (eg target and background).
Colour coding
Colour coding is a process by which different colours are
used to represent different categories of information. For
example, a red signal means ‘Stop’ while a green signal means
‘Proceed’. If colour provides a completely unique source of
information, the coding is called non-redundant. Colour can
also be combined with other coding dimensions such that
two or more codes correlate with one another, eg hand
signalling arrangements. This is referred to as redundant
coding.
Page 179
Part 3: Reference
Jargon buster
Competence
The orchestration of the knowledge, skills and attitudes
required to perform a specific range of tasks, job or role to a
prescribed standard.
Degraded operating conditions
State of continuing railway operations with significant
equipment failures (such as track related failures or
communication system failures).
Competence assessment
A management process designed to ensure that staff have
the knowledge, skills and attitudes necessary to perform
their work to the standard expected.
Design scenario analysis
Design scenario analysis – see Techniques
Concentrator
A telephonic device used on the railways (eg in signal boxes)
that ensures only one phone call from potentially many
sources (land-line, SPT, mobile) can be received at one time
– thus reducing attentional demand.
Context-Sensitive Help
Help in which the help text or range of help users is derived
from the contextual information associated with the user’s
last input, selected object, or the current location within the
system or application.
CRM
Crew Resource Management – an aviation industry
programme designed to assure the quality of teamworking
amongst all the crew of an aircraft.
CUD
Dialogue
An interaction between a user and an interface to achieve a
particular goal.
Diffuse lighting
Lighting which comes from many directions, none of which
predominates.
Disability
Any restriction or lack (resulting from an impairment) of
ability to perform an activity in the manner or within the
range considered normal for a human being.
Disability glare
Glare produced directly or by reflection that impairs the
vision of objects without necessarily causing discomfort.
Discrimination
The detection of a just noticeable difference between stimuli
(eg colour differences between visual stimuli).
Effectiveness
The relationship between the output produced and
the outcome intended: it is a measure of how far a preexisting situation is improved by the product of a resource.
Alternatively, it is a measure of how far the product of the
resource has achieved a pre-specified goal.
Efficiency
The relationship between resources and output: it is a
measure either of how fast the resource produces a
required output, or how little resource is required.
Emergency operating conditions
State of the railway in response to a major safety or
security-related event. Introduction of contingency plans at a
moment’s notice.
Ergonomics
The study of human capabilities and limitations, human
interaction with technologies and environments, and the
application of this knowledge to products, processes and
environments.
Error
A mismatch between the user’s goal and the response of
the system or environment of which the user is part. Errors
can include navigation errors, syntax errors, and conceptual
errors.
Communications Usage Diagram - see Techniques.
Culture
The set of values and norms that govern how people
understand – and what they expect from – each other
within an organisation. Culture is both an input and output
of behaviour. It determines and facilitates it as well as
emerging from, and changing (albeit slowly) as a result of it.
Page 180
Display resolution
The number of separately addressable pixels on a display
screen.
Economy
The relationship between cost and resources: it is the cost
by which a resource produces a required output.
Fault trees
Fault trees – See Techniques.
Feedback
Output presented by the interface in reaction to the user’s
input.
Understanding Human Factors/June 08
Part 3: Reference
Jargon buster
Fitness for purpose
The capability of a product to serve the purpose for which it
was designed, in the way it was designed to be used.
FOC
Freight Operating Company.
Focus groups
Focus groups – see Techniques.
Front-end analysis
Collective term for those analyses conducted at the earliest
stages of system design and concerned with a system’s
personnel, training and logistics requirements.
Glare
The discomfort or impairment of vision experienced when
parts of the visual field are excessively bright in relation to
the general surroundings.
Groupware Task Analysis
See Techniques.
HE HAZOP
Human Error HAZard Operability - see Techniques.
HEART
Human Error Assessment and Reduction Technique - see
Techniques.
Function allocation
The process of deciding how system functions should be
assigned to the human and machine elements of a system.
Fixed function allocation means that these decisions are
made at design time and remain fixed for the system’s life.
In this case, function allocation is the most basic of system
design decisions since it establishes the framework within
which the design of the system (equipment, workspace,
training, etc) is developed. Dynamic function allocation takes
place during the task. Between people and machines, it is
known as adaptive automation; between people it is known
as teamworking.
HEP
Human Error Probability.
Function analysis
The analysis of system functions. Functions describe activities
which may be implemented by personnel alone (control a
signalling section), by equipment alone (self-test/equipment
circuitry), or, as in most cases, by some combination of both
(pre-flight checks). Functions can be instantaneous (set
route) or prolonged (monitor panel), simple (accelerate) or
complex (assess emergency situation).
HFACS
Human Factors Analysis and Classification System. A
comprehensive method of root cause analysis used to
identify and classify the human causes of incidents.
Understanding Human Factors/June 08
Heuristic analysis
Heuristic analysis – see Techniques.
HFIP
Human Factors Integration Plan. A plan that defines the
integration of human factors into system development and
appropriate assurance procedures to ensure that these
activities are completed.
HRA
Human Reliability Analysis - see Techniques.
HRO
High Reliability Organisation. This is a safety-critical
organisation which has far fewer safety incidents than might
be expected due to its staff having learned to manage the
unexpected – mainly though sensitivity to error and decision
making that depends on expertise rather than hierarchy.
HSE
Health and Safety Executive.
HTA
Hierarchical Task Analysis - see Techniques.
Hertz (Hz)
SI unit of frequency, indicating the number of cycles per
second (c/s).
Hexagons
Hexagons – see Techniques.
Hue
The term that most closely resembles our notion of ‘colour’,
for example, red, green and blue. It is that quality of a colour
that cannot be accounted for by brightness or saturation
differences. An objective measure of hue is provided by
the dominant wavelength of that colour’s spectral power
distribution.
Human factors
All of the ‘people’ issues that must be considered to
assure the lifelong safety and effectiveness of a system or
organisation.
Page 181
Part 3: Reference
Jargon buster
Human interface
The controls, devices and displays which an operator uses to
control, monitor, or otherwise interact with, the rest of the
system. Also known as the Man-Machine Interface (MMI).
Icon
A pictorial representation consisting of an image, with or
without a label, presented on a display screen, designed to
provide a visual reminder of the name, structure, appearance
or purpose of the entity that it represents.
IECC
Integrated Electronic Control Centre. A computer
workstation equipped with a series of software-animated
vdu-based diagrams used by signallers to track the progress
and status of trains in a specified geographical area under
their control.
Interface surveys
Interface surveys – see Techniques.
Interview
Interview – see Techniques.
IPME
Integrated Performance Modelling Environment - see
Techniques.
ISA
Instantaneous Self-Assessment (of workload ) – see
Techniques.
Job
The grouping of tasks, roles and responsibilities constituting
the principal work assignment of one person.
Page 182
KLM
Keystroke Level Model - see Techniques
Knowledge
Knowledge constitutes a representation of something
in a way that the knower can utilise for some purpose.
This representation may or may not be articulable and
the purpose for which it is used may or may not be
predetermined. This allows for the fact that knowledge can
be used innovatively and inductively (re-purposed).
KSA
Knowledge, skills and attitudes.
Layout Analysis
Layout analysis – see Techniques.
Lifecycle
The development, operation and maintenance of a system,
spanning its life from the definition of its requirements to its
disposal following the termination of its use.
Link analysis
Link analysis – see Techniques.
Methodology
An integrated set of tools, techniques and procedures,
collectively aimed at a specified goal.
Maintainability
The ability to carry out rapid and reliable system restoration,
using people trained to a specified level, and specified
support facilities to maintain the equipment at a specified
level of performance.
Maintenance
All activities necessary to keep materiel in, or restore it to, a
specified condition.
Morale
A state of mind that at one extreme can allow a group of
people to commit to a shared objective with complete
disregard for the potential cost to themselves. At the other
extreme it can account for little or no effective or organised
action in the face of clear situational demands.
Motivation
A state of mind that releases the necessary emotional
energy for an individual to commit themselves to act
effectively towards a declared objective.
Murphy Diagrams
Murphy diagrams – see Techniques.
NASA – TLX
NASA Task Load Index – see Techniques.
Normal operating conditions
State of continuing railway operations according to normal
timetables, incorporating minor disturbances and delays to
the service in traffic hours and operations in non-traffic
hours.
NX panel
eNtry eXit panel. A type of electrically operated display
panel set out as a diagram and used by signallers to track
the progress and status of trains and set their routes in a
specified geographical area under their control. NX panels
are equipped with push-pull switches for route setting
and other switches to control point settings, level crossing
cameras etc.
Understanding Human Factors/June 08
Part 3: Reference
Jargon buster
Observational analysis
Observational analysis – see Techniques.
Performance appraisal
A management process designed to ensure that staff are
motivated in their jobs and are developed to their full
potential.
Pixel
The smallest addressable display element that is capable of
generating the full colour and/or luminance ranges of the
display.
Product
Any equipment, process, procedure, rule or instruction that
has been designed to serve some pre-defined purpose.
Prototype
Any artefact created for the purpose of demonstration to
users in order to elicit or test user feedback. This includes
demonstrators, mock-ups, paper prototypes, simulations,
role-plays, dummy systems or documents, and scenarios.
Psychometric test
A test of a specific mental ability or process, eg spatial ability,
numerical reasoning, critical thinking.
Questionnaire
Questionnaire – see Techniques.
QUIS
Questionnaire for User Interface Satisfaction - see Techniques.
Understanding Human Factors/June 08
Responsibility
A set of tasks and duties for which an individual is
accountable in terms of their conduct, performance and
management.
Risk
An estimate of the propensity for an otherwise stable
situation to suddenly fail. The estimate may be a formal
one, based on a risk assessment methodology, or it may be
entirely intuitive, based on subjective feelings. The presence
of perceived risk in a situation is often fundamental to its
interest for humans. Without it, people become bored
and/or indulge in behaviour to increase the risk level. The
perceived level of risk in a situation can be affected by a
large range of different factors.
Role
A set of expectations placed upon an individual by an
organisation and realised by that individual through the
execution of their job.
RSSB
Rail Safety and Standards Board.
SA
Situation awareness. The ability to know what is going on
around you, and to use this information to project accurately
into the future, resulting in successful plans for predicted
events and their contingencies.
Safety culture
The set of values and priorities placed on all aspects of
safety by everyone at every level of an organisation.
SAGAT
Situation Awareness Global Assessment Technique. A
simulation-based technique in which the task is stopped
periodically so that the user can be asked about their
perception of the situation at that instant.
SART
Situation Awareness Rating Technique - see Techniques.
SASHA
Situational Awareness for SHAPE – see Techniques.
Saturation
The quality that distinguishes a hue from white. Pastel shades
are de-saturated, vivid colours are saturated. An objective
measure of saturation is purity.
SHAPE
Solutions for Human-Automation Partnerships in
European (Air Traffic Management Systems).
SHERPA
Systematic Human Error Reduction and Prediction Approach
- see Techniques.
Skill
An organised and co-ordinated pattern of mental and/or
physical activity, which becomes more accomplished with
training or other experience.
Skill fade
The tendency of learned skills to degrade without sufficient
practice. Different skills fade at different rates and are
affected by multiple factors.
Page 183
Part 3: Reference
Jargon buster
SME
Subject Matter Expert. A person with task knowledge, skills
and accepted qualification in a specific domain.
SUMI
Software Usability Measurement Inventory – see Techniques.
SUS
SPAD Hazard Checklist
A tool developed by RSSB to identify working practices and
communications procedures that should be adopted by both
signallers and drivers to help prevent SPADs occurring.
Staffing
A continuous process by which an organisation arranges
for appropriate numbers of people with the appropriate
qualities to be available, so that it may operate its business
safely and effectively.
Stakeholder
Any individual who is affected by the output from, provides
the input to, develops, maintains, uses or manages the use of
a system or product.
Stress
An adverse reaction people have to excessive arousal or
pressure. Stress is not a disease, but if it goes on unchecked,
it can lead to mental and physical ill-health.
Stressor
An impelling force which produces a demand upon physical
or mental energy.
SSM
Soft Systems Methodology - see Techniques.
Style guide
A document which sets out the design principles, rules and
conventions agreed by all of the stakeholders.
Page 184
System Usability Scale - see Techniques.
SWAT
Subjective Workload Assessment Technique - see Techniques.
System
Any set of elements, including physical equipment, computer
software, human users and procedures, interacting and
organised in relation to a goal.
Task analysis
The elicitation and representation of a set of tasks in order
to understand the relationships between their constituent
activities, performance criteria and objectives.
Task synthesis
The process of specifying and putting together the tasks of
which a system function consists.
TCSD
Task-Centred System Design - see Techniques.
TCTA
Team Cognitive Task Analysis - see Techniques.
Systems analysis
A generic term for the various analytic techniques applied
before or during the system design stage, eg requirements
analysis, function analysis, front end analysis etc.
Systems engineering
The processes by which system requirements are developed
into a system performance, design, and production
specification; and by which that specification is then
transformed into a fabricated, prototyped, integrated and
tested product.
TAD
Target Audience Description. A descriptive profile of the
characteristics, skills and abilities of the ‘end-user’ of a
designed system.
Task
The set of physical and mental interactions that are required
within a work environment in order to achieve a prescribed
goal.
TDRE
Team Decision Requirements Exercise - see Techniques.
Teamworking
A team is a set of two or more individuals who interact
adaptively with each other to achieve specified, shared and
valued objectives. The period of interaction may be anything
from a few seconds (eg a driver and signaller) to many years
(eg a large infrastructure project team). Teamwork refers to
the activities performed by team members in response to
each other’s needs and expectations, to enable the team to
achieve its shared objectives.
TIP
Teamworking Improvement Process - see Techniques.
TNA
Training Needs Analysis.
TOC
Train Operating Company.
Understanding Human Factors/June 08
Part 3: Reference
Jargon buster
Training
A continuous process by which an organisation arranges for
appropriate knowledge, skills and attitudes to be available to
its workforce, so that it may operate its business safely and
effectively.
TRUST
Train Running System TOPS. Network Rail’s system
for monitoring the punctuality of trains, mainly fed by
automatic inputs from signalling systems.
TTRAM
Task and training requirements methodology - see
Techniques.
TWA
Team Workload Assessment - see Techniques.
Usability
The ability of a product to be understood and operated
efficiently, safely and effectively by its intended users.
Usability also includes how well the use of the product fits
with related products and their users within and across
organisations.
User
Anyone who employs an artefact to carry out a task.
User-centred
Approaches (generally to design) which have as their
primary focus the consideration of the interests of the
individuals who will work with, or use the output from, a
piece of equipment.
Validation (internal)
The ability of a test, rule, procedure or system to achieve the
objectives set for it. For example, a selection test is internally
valid if it successfully discriminates between the people it
sets out to discriminate between; a system is internally valid
if it successfully implements all of the specifications drawn up
for it.
Validation (external)
The ability of a test, rule, procedure or design to achieve
operational objectives. For example, a selection test is
externally valid if the people it predicts will do well actually
do so; a system is externally valid if it allows the organisation
to achieve the operational objectives set out for it. It is
possible for something to be highly internally valid (ie it does
exactly what it is supposed to) but of low external validity
(ie what it does has little value in an operational setting).
Visual acuity
The capacity for discriminating between objects which are
very close together. The expression more commonly used
for an individual’s visual acuity is the ratio of the distance at
which the individual can read a line on a standard optician’s
chart to the standard distance at which a person of normal
sight can read that line (eg 6/12 means that the individual
can just read at 6 m the line which a normally sighted
person can just read at 12 m).
Visual impairment
Any loss or abnormality of psychological, physiological or
anatomical structure or function relating to vision.
Visual field
The full extent in space of what can be seen when looking in
a given direction.
VDU
Visual Display Unit (typically, a computer screen).
Walk-through Analysis
Walk-through analysis – see Techniques.
Verification
The process of ascertaining that a test, rule, procedure or
system does what its designers intended. Verification is the
process needed to demonstrate internal validity.
Why-because Analysis
Why-because analysis – see Techniques.
Virtual environment
An artificial environment generated by an immersive display
(eg goggles) which presents information in such a way
as is appropriate to give the operator the perception of
viewing and interacting with objects in three-dimensional
surroundings.
Workload
The effort demanded from people by the tasks they have to
do. It can be the effort demanded at a single point in time,
or over a whole shift. Workload can be physical or mental.
It can be the physical demands created by working in a
particular posture, manual labour or working in particular
environmental conditions; or it can be the mental demands
created by the need to attend to sources of information and
then process the information – often against time pressure.
User trial
User trial – see Techniques.
Understanding Human Factors/June 08
Page 185
Part 3: Reference
Jargon buster
Workplace
The complete working environment within which the
operator(s) and equipment(s) are arranged to function as a
unit.
Workshop
Workshop – see Techniques.
Page 186
Understanding Human Factors/June 08
understanding
Bibliography
human factors
understanding
Bibliography
human factors
Part 3: Reference
Bibliography
Bibliography
This bibliography includes all of the references listed at
the end of the sections in Part 2. It has been extended
with many other references that were used in the
development of this Guide and by other material that
may be of use to readers who wish to delve deeper into
any of the issues or techniques presented in these pages.
8
9
Annett J. & Duncan K.D. (1967) Task Analysis and
training design, Occupational Psychology, 41, 211221
Annett J. Duncan K.D. Stammers R.B. & Gray M.J.
(1971) Task Analysis. Training Information No. 6.
HMSO: London
1
Adair J. (2002) Effective Leadership, Pan
10 Annett J. & Stanton N.A. (2000) Task Analysis, Taylor
& Francis
2
Adair J. (2004) The John Adair Handbook of
Management and Leadership, Thorogood
11 Armstrong M. & Stephens T. (2005) A Handbook of
Management and Leadership, Kogan Page
3
Adair J. (2004) Adair on Teambuilding and
Motivation, Thorogood
12 ASLEF Booklet Shift work, lifestyle and health
4
Ainsworth L.K. & Marshall E. (1998) Issues of quality
and practicality in task analysis: preliminary results
from two surveys. Ergonomics 41(11), 1604-1617.
Reprinted in J. Annett & N.A. Stanton (2000) op.cit.
Pp. 79-89
5
6
7
Air Affairs Ltd (2003) Safety critical roles: refresher
training and re-assessment frequency, RSSB
Research report, Rserv479. Available at www.rssb.
co.uk/pdf/reports/research/safety%20critical%20
roles%20-%20refresher%20training%20and%20reassessment%20frequency.pdf (as of May 2008)
ANEP 27 (NATO MAS), Edition 1. Human Factors
guidelines for the design of MMIs in Operational
Rooms
Annett J. Cunningham D.J. & Mathias-Jones P. (2000)
A method for measuring team skills. Ergonomics,
43(8), 1076-1094
Understanding Human Factors/June 08
13 Baber C. & Stanton N.A. (1996) Human error
identification techniques applied to public
technology: predictions compared with observed
use. Applied Ergonomics, 27(2), 119-131
14 Baber C. & Stanton N.A. (1994) Task analysis for
error identification. Ergonomics 37, 1923-1941
15 Baber C. & Stanton N.A. (1996) Observation as a
technique for Usability Evaluations. In P. Jordan et al
(eds.), Usability in Industry, 85-94, Taylor & Francis
16 Bailey R.W. (1982) Human Performance Engineering:
A Guide for System Designers, Prentice-Hall, NJ
17 Bainbridge L. (1987) New Technology and Human
Error, chapter 24, pages 271-283, John Wiley
18 Bass A. Aspinal J. Walter G. & Stanton N.A. (1995) A
software toolkit for hierarchical task analysis. Applied
Ergonomics. 26 (2) pp. 147-151
19 Bennis W. & Nanus B. (1985) Leaders: Strategies for
Taking Charge, Harper Collins
20 Boff K.R. & Lincoln J.E. (1988) Engineering Data
Compendium: Human Perception and Performance.
John Wiley
21 Boyce P.R. (1997) Illumination. In Handbook of
Human Factors and Ergonomics, Edited by G.
Salvendy, (2nd Ed), John Wiley
22 Braband J. & Brehmke B. (2002) Application of
Why-Because Graphs to Railway Near-Misses, In
Workshop on the Investigation and Reporting of
Accidents 2002, (IRIA 2002) Ed C. W. Johnson.
GIST Technical Report F2002-2, Department Of
Computing Science, University Of Glasgow, Scotland
23 Briault S. (1994) What is integrated learning? Training
Officer, May, Vol. 30, No. 4
24 BPS psychological testing website, British
Psychological Society
www.psychtesting.org.uk/ (as of May 2008)
25 Broadbent D.E. (1958) Perception and
Communication. Pergamon Press, London
26 Broadbent D.E. (1979) Effects of Noise on
Behaviour. In: C M Harris (ed) Handbook of noise
control. 2nd edition. McGraw Hill, New York
27 BS 4727: PART 3: GROUP 08: (1995) Glossary
of Electrotechnical power, telecommunication,
electronics, lighting and colour terms. Terms
particular to telecommunications and electronics.
Acoustics and electro-acoustics
Page 189
Part 3: Reference
Bibliography
28 BS 5266-1 (1999) Emergency Lighting. Code of
Practice for the emergency lighting of premises
other than cinemas and certain other specified
premises used for entertainment
29 BS 5378-2 (1980) Safety signs and colours.
Specification for colorimetric and photometric
properties of materials
30 BS 5499-1 (2002) Graphic symbols and signs. Safety
signs, including fire safety signs. Specification for
geometric shapes, colours and layouts
37 BS 5940 Part 1 (1980) Design and dimensions of
office workstations, desks, tables and chairs
38 BS 6548-2 (1992) IEC 60706-2 (1990)
Maintainability of equipment. Guide to
maintainability studies during the design phase
39 BS 7000-2 (1997) Design management systems.
Guide to managing the design of manufactured
products
40 BS 7324 (1990) ISO 7000 (1989) Guide to
graphical symbols for use on equipment
47 BS EN 547-2 (1997) Safety of machinery. Human
body measurements. Principles for determining the
dimensions required for access openings
48 BS EN 614-2 (2000) Safety of machinery. Ergonomic
design principles. Interaction between the design of
machinery and work tasks
49 BS EN 842 (1996) Safety of machinery. Visual
danger signals. General requirements, design and
testing
50 BS EN 894-1 (1997) Safety of machinery.
Ergonomics requirements for the design of displays
and control actuators
31 BS 5499-2 (1986) Fire safety signs, notices and
graphic symbols. Specification for self-luminous fire
safety signs
41 BS 7517 (1995) IEC 61227 (1993) Nuclear power
plants. Control rooms. Operator controls
32 BS 5499-3 (1990) Fire safety signs, notices and
graphic symbols. Specification for internally
illuminated fire safety signs
42 BS EN 292-1 (1991) Safety of machinery. Basic
concepts, general principles for design. Basic
terminology, methodology
33 BS 5499-4 (2000) Safety signs, including fire safety
signs. Code of practice for escape route signing
43 BS EN 292-2 (1991) Safety of machinery. Basic
concepts, general principles for design. Technical
principles and specifications
34 BS 5499-5 (2002) Graphic symbols and signs. Safety
signs, including fire safety signs. Signs with specific
safety meanings
44 BS EN 349 (1993) Safety of machinery. Minimum
gaps to avoid crushing of parts of the human body
54 BS EN ISO 266 (1997) Acoustics. Preferred
frequencies
35 BS 5499-6 (2002) Graphic symbols and signs. Safety
signs, including fire safety signs. Creation and design
of graphical symbols for use in safety signs
45 BS EN 457 (1992) Safety of machinery. Auditory
danger signals. General requirements, design and
testing
55 BS EN ISO 7029 (2000) Acoustics. Statistical
distribution of hearing thresholds as a function of
age
36 BS 5817-8 (1990) IEC 60574-16 (1987)
Specification for audio-visual, video and television
equipment and systems
46 BS EN 547-1 (1997) Safety of machinery. Human
body measurements. Principles for determining the
dimensions required for openings for whole body
access into machinery
56 BS EN ISO 9241 (1996-2000) Ergonomic
Requirements for Office Work with Visual Display
Terminals.
Page 190
51 BS EN 981 (1997) Safety machinery. System of
auditory and visual danger and information signals
52 BS EN 60598-1 (2000) Luminaires. General
Requirements and Tests
53 BS EN 60651 (1994) Specification for Sound Level
Meters
Understanding Human Factors/June 08
Part 3: Reference
Bibliography
57 BS EN ISO 9241-1 (1997) Ergonomic Requirements
for Office Work with Visual Display Terminals
(VDTs) Part 1 General Introduction
66 BS EN ISO 9241-13 (1999) Ergonomic
requirements for office work with visual display
terminals. User Guidance
76 BS EN ISO 13406-1 (1999) Ergonomic
requirements for work with visual displays based on
flat panels – Introduction
58 BS EN ISO 9241-2 (1992) Ergonomic requirements
for office work with visual display terminals.
Guidance on task requirements
67 BS EN ISO 9241-14 (2000) Ergonomic
requirements for office work with visual display
terminals. Menu dialogues
59 BS EN ISO 9241-3 (1992) Ergonomic requirements
for office work with visual display terminals. Visual
display requirements
68 BS EN ISO 9241-15 (1998) Ergonomic
requirements for office work with visual display
terminals. Command dialogues
77 BS EN ISO 13406-2 (2002) Ergonomic
requirements for work with visual displays based on
flat panels. Ergonomic requirements for flat panel
displays
60 BS EN ISO 9241-4 (1998) Ergonomic requirements
for office work with visual display terminals.
Keyboard requirements
69 BS EN ISO 9241-16 (1999) Ergonomic
requirements for office work with visual display
terminals. Direct manipulation dialogues
61 BS EN ISO 9241-6 (2000) Ergonomic requirements
for office work with visual display terminals.
Environmental requirements
70 BS EN ISO 9241-17:1998. Ergonomic requirements
for office work with visual display terminals. Form
filling dialogues
62 BS EN ISO 9241-7 (1998) Ergonomic Requirements
for Office Work with Visual Display Terminals
(VDTs) Part 7 Display Requirements with
Reflections
71 BS EN ISO 10075-1 (2000) Ergonomic Principles
Related to Mental Workload. General terms and
definitions
63 BS EN ISO 9241-8 (1998) Ergonomic Requirements
for Office Work with Visual Display Terminal (VDTs)
Part 8 Requirements for Displayed Colours
64 BS EN ISO 9241-11 (1998) Ergonomic
requirements for office work with visual display
terminals. Guidance on usability
65 BS EN ISO 9241-12 (1999) Ergonomic
requirements for office work with visual display
terminals. Presentation of information
Understanding Human Factors/June 08
72 BS EN ISO 10075-2 (2000) Ergonomic Principles
Related to Mental Workload. Design Principles
73 BS EN ISO 11064-1 (2001) Ergonomics Design
of Control Centres. Principles for the Design of
Control Centres
74 BS EN ISO 11064-2 (2001) Ergonomics Design of
Control Centres. Principles for the Arrangement of
Control Suites
75 BS EN ISO 11064-3 (2001) Ergonomics Design of
Control Centres. Control Room Layout
78 BS EN ISO 13407 (1999) Human-centred design
processes for interactive systems
79 The ISO 13407 standard provides a framework
for user centred development activities that can be
adapted to numerous development environments:
from a straight waterfall type of development
process to an iterative type of environment
80 BS EN ISO 14915-1 (2002) Software ergonomics
for multimedia user interfaces – Part 1: Design
principles and framework
81 BS EN ISO 14915-3 (2002) Software ergonomics
for multimedia user interfaces – Part 3: Media
selection and combination
82 BS ISO 2631-4 (2001) Mechanical vibration and
shock. Evaluation of Human Exposure to Wholebody Vibration. Guidelines for the evaluation of
the effects of vibration and rotational motion on
passengers and crew comfort in fixed-guideway
transport systems
83 BS ISO 3864-1. Safety colours and safety signs. Part
1. Safety signs in workplaces and public areas. Design
principles (ISO 3864-1) system interfaces and
Page 191
Part 3: Reference
Bibliography
symbols – Icon symbols and functions Part 6: Action
icons 99/561489 DC
84 BS ISO 7731 (2001) Ergonomics: Danger signals for
public and work places – Auditory danger signals
85 BS ISO 9186 (2001) Graphical symbols. Test
methods for judged comprehensibility and for
comprehension
86 BS ISO 9921-1 (1996) Ergonomic assessment
of speech communication. Speech interference
communication distances for persons with normal
hearing capacity in direct communication (SIL
Method)
87 BS ISO 9921 Ergonomics. Assessment of speech
communication 01/564713 DC
88 BS ISO/IEC 15288 (2002) Systems engineering
– System life cycle processes.
93 Cha D.W. (2001) Comparative study of subjective
workload assessment techniques for the evaluation
of ITS-orientated human-machine interface systems.
Journal of Korean Society of Transportation. Vol 19
(3), 45-58
94 Checkland P. (1981, 1998) Systems Thinking, Systems
Practice, John Wiley
95 Checkland P. (1989) Soft systems methodology,
in ROSENHEAD J (Ed.), Rational analysis for a
problematic world, 71-120, John Wiley
96 Checkland P. & Scholes J. (1990) Soft systems
methodology in action, John Wiley
97 Chin J.P. Diehl V.A. & Norman K.L. (1988)
Development of an instrument Measuring User
Satisfaction of the Human-Computer Interface.
CHI’88
89 Buckley R. & Caple J. (1992) The Theory and
Practice of Training, Kogan Page
98 Ciavarelli A. (2002) Human Factors Checklist:
An Aircraft Accident Investigation Tool, School of
Aviation Safety, California. Internet source
90 Card S.K. Mackinlay J.D. & Schneiderman B. (1999)
Readings in Information Visualization: Using Vision to
Think: Morgan Kaufmann
99 CIBSE Code for Interior Lighting (1994) Chartered
Institution of Building Service Engineers
91 Card S.K. Moran T.P. & Newell A. (1983) The
Psychology of Human Computer Interaction.
Hillsdale, NJ. Erlbaum L. (Reference for KLM)
92 Carroll, J.M. (1990). The Nurnberg Funnel. Designing
Minimalist Instruction for Practical Computer Skill.
Cambridge, MA: The MIT Press, 1990
Page 192
103 Coleman G.J. (1979) Determination of sound
pressure levels for acoustic warning signals in quiet
conditions. Institute of Occupational Medicine,
Edinburgh: Ergonomics Branch Report ER/79/01
104 Coleman G.J. et al (1984) Communications in noisy
environments. Institute of Occupational Medicine,
Edinburgh, TM84
105 Covey S.R. (1999) The Seven Habits of Highly
Effective People, Simon & Schuster
106 Crainer S. & Dearlove D. (Editors) (2004) Financial
Times Handbook of Management, Financial Times
Prentice Hall
107 Cronbach L.J. & Gleser G.C. (1965) Psychological
tests and personnel decisions, Urbana, University of
Illinois Press
108 Cushing S. (1994) Fatal Words: Communication
Clashes and Aircraft Crashes, University of Chicago
Press
109 Darnell M. (1996-2004) www.baddesigns.com (for
details on commonplace examples of design failure)
100 CIBSE Lighting Guide 3 (LG3) (1996) The visual
environment for display screen equipment use
110 Davies D. (2000) Automatic Train Protection for
the Railway Network in Britain – A study. RA Eng.
London
101 Clark H. (1996) Using Language, Cambridge
University Press
111 Davis M. et al. (2003) The Relaxation and Stress
Reduction Workbook, New Harbinger Publications
102 Code of Practice: Competence Assessment, GO/
RC3560 (1998) Safety & Standards Directorate,
Railtrack PLC, Floor DP01, Railtrack House, Euston
Square, London NW1 2EE
112 Deal T. & Kennedy A. (1988) Corporate Cultures:
The Rites and Rituals of Corporate Life, Penguin
Understanding Human Factors/June 08
Part 3: Reference
Bibliography
113 Dean T.F. (1997) Directory of Design support
methods, Defence Technical Information Centre,
DTIC-AM. MATRIS Office, ADA 328 375,
September
114 Def Stan 02-556 (2001) Guide to design and user
requirements of internal communication equipment,
MOD
115 Def Stan 02-599 (2001) Policy requirements and
design guidance for alarms and warning systems,
MOD
116 Def Stan 00-25 Human Factors for Designers of
Systems Part 15 Issue 1, Principles and Processes,
MOD, www.dstan.mod.uk/data/00/025/15000100.
pdf (as of May 2008)
development programs; Proceedings of the Human
Factors Society, 30th Annual Meeting 1014-1018
123 Doll T.J. & Folds D.J. (1986) Auditory signals in
military aircraft: ergonomics principles versus
practice. Appl Erg 17(4) 257-264
124 Drucker P. (1999) Practice of Management,
Butterworth Heinemann
125 Drury C.G. (1990) Methods for direct observation
of performance, In Wilson J. & Corlett E. N.
Evaluation of Human Work: A practical Ergonomics
Methodology, 2nd Edition, Taylor & Francis, 45-68
126 Du Ross S.H. (1978) Civil aircraft warning systems
– a survey of pilot opinion within British Airways.
Royal Aircraft Establishment. Tech Rpt 78056
117 Def Stand 00-25 Human Factors for Designers of
Systems Part 19: Technical Guidance and Data –
Human Engineering Domain, MOD
127 Dukes C. (2001) Easy Step by Step Guide to
Recruiting the Right Staff, Rowmark
118 Dekker S. (2002) The field guide to human error
investigations. Aldershot: Ashgate
128 Dumas S.J. & Redish J.C. (1993) A practical guide
to usability testing, Norwood: Ablex Publishing
Corporation
119 Department for International Development (DFID)
(1995), Enhancing stakeholder participation in aid
activities, Technical Note
120 Diaper D. & Stanton N.A. (2003) The Handbook
of Task Analysis for Human Computer Interaction.
Lawrence Erlbaum Assoc Inc
121 Dietrich R. (Ed) (2003) Communication in High Risk
Environments, Helmut Buske Verlag
122 Ditzian J.L. Purifoy G.R. Sullivan G.K. & Bogner M.S.
(1986) Embedded training: lessons from system
Understanding Human Factors/June 08
129 Dupont, G. (1997) The Dirty Dozen Errors
in Maintenance. In: proceedings of the 11th
Symposium on Human Factors in Aviation
Maintenance
130 Eason K. (1987) Information technology and
organizational change, Taylor & Francis
131 Easterby R. (1984) Tasks, processes and display
design. In R. Easterby & H. Zwaga (Eds.), Information
Design (pp. 19-36) John Wiley
132 Eden C. (1989) Using cognitive mapping for
strategic options development and analysis
(SODA), in Rosenhead J (Ed.), Rational analysis for a
problematic world, 21-70, John Wiley
133 Edkins G.D. & Pollock C.M. (1997) The influence of
sustained attention on railway accidents. Accid. Anal.
& Prev. Vol 29, No. 4, pp553-539, Elsevier Science
134 Embrey D.E. (1986) SHERPA: A systematic human
error reduction and prediction approach. Paper
presented at the International Meeting on Advances
in Nuclear Power Systems, Knoxville, Tennessee
135 Embrey D.E. (1993) Quantitative and qualitative
prediction of human error in safety assessments.
Institute of Chemical Engineers Symposium Series,
130, 329-350
136 Ernst & Young (2000) Development of a leadership
resource pack, Offshore Technology Report
2000/098, HSE
137 Fisher R. & Ury W. (1999) Getting to Yes:
Negotiating an agreement without giving in. Random
House
138 Fitt. P.M. (ed) (1951) Human engineering for an
effective air-navigation and air traffic control system.
Colombus Ohio: Ohio State University Research
Foundation
139 Flaherty, M (2002) Rural Institute article on staff
retention, University of Montana, http://ruralinstitute.
umt.edu/training/publications/newsletters/staff_
retention.asp (as of May 2008)
Page 193
Part 3: Reference
Bibliography
140 Flanagan J.C. (1954) The Critical Incident Technique.
Psychological Bulletin, 51, 327-358
Handbook of Task Analysis for Human Computer
Interaction. Lawrence Erlbaum Associates
141 Fleming M. & Lardner R. (1999) Safety Culture: The
way forward, The Chemical Engineer, 11 March 1999
151 Greener M. (2003) The Which? Guide to Managing
Stress, Which? Books
142 Fleming M. & Lardner R. (2000) Behaviour
Modification Programmes: Establishing Best Practice,
The Keil Centre, Edinburgh. Research performed for
the HSE
152 Greenstreet Berman Ltd (2003) Competence
assessment for the hazardous industries. Research
Report 086 Health and Safety Executive
143 Fletcher C. (2004), Appraisal and Feedback: Making
performance review work, Chartered Institute of
Personnel & Development
144 Flowers K. (1975) Handedness & Controlled
Movement. British Journal of Psychology, 66, 39-52
145 Fournies F. (1999) Coaching for Improved Work
Performance, McGraw-Hill Education
146 Gagné R.M. Briggs L.J. & Wager W.W. (1992)
Principles of Instructional Design. Holt, Rinehart &
Winston
147 Go K. & Carroll J.M. (2003) Scenario-Based Task
Analysis. In D. Diaper & N. Stanton (Eds) The
Handbook of Task Analysis for Human-Computer
Interaction, Lawrence Erlbaum Associates
148 Goldstein I.L. & Ford, K. (2001) Training in
Organisations, Wadsworth
149 Goleman D. (1998) Working with Emotional
Intelligence, Bloomsbury
150 Greenberg S. (2003) Working through Task-Centred
System Design. In D. Diaper & N. Stanton (Eds) The
Page 194
153 Gregory D. & Shanahan P. (2004) Teamworking best
practice in the railway industry: The Journey Guide,
Gregory Harland Ltd, for RSSB, Euston
154 Gregory Harland Ltd, (1999) Development of a
skill fade model, GHL/CHS/SkillFade/Deliverables/
FinalReport/Volume1/v3.0, Centre for Human
Sciences, DERA
160 Health and Safety Commission. (1992) Display
Screen Equipment Work, Health and Safety (Display
Screen Equipment) Regulations 1992, Guidance on
Regulations L26
161 Heil G. Douglas McGregor (2000) Revisited Managing the Human Side of the Enterprise, John
Wiley
162 Helmreich R.L. & Merritt A.C. (1998) Culture
at Work in Aviation and Medicine: National,
Organizational and Professional Influences, Ashgate
163 Hendrick, K. and Benner, L. (1987) Investigating
accidents with STEP. New York: Marcel Decker
164 Herzberg, F. Mausner, B. & Snyderman, B.B. (1959)
The Motivation to Work (2nd ed.), John Wiley
155 Handy C. (1993) Understanding Organisations, 4th
Edition, Penguin
165 Herzberg F. (1993) Motivation to Work, Transaction
Publishers
156 Handy C. (1999) Inside Organisations – 21 Ideas for
Managers, Penguin
166 HM Inspectorate of Constabulary (1999), Managing
Learning: A Study of Police Training, HMIC, 50
Queen Anne’s Gate, London, SW1H 9AT: ISBN
1840 82 2716
157 Hansen, D. see http://iso9k1.home.att.net/pa/
performance_appraisal.html (as of May 2008)
158 Hart S.G. & Staveland L.E. (1988) Development of
a multi-dimensional workload rating scale: Results of
empirical and theoretical research. In P. A. Hancock
& N. Meshkati (Eds.), Human Mental Workload,
Elsevier, Amsterdam
159 Harvard Business Review (2003) on Motivating
People, Harvard Business School Press
167 Hodgson A.M. (1992) Hexagons for systems
thinking, European Journal of Operational Research,
59, 220-230
168 Hofstede G. (1994) Cultures and Organisations
– Intercultural Cooperation and its Importance for
Survival, HarperCollins Business
169 Hollnagel E. (1993) Human Reliability Analysis:
context and control, Academic Press
Understanding Human Factors/June 08
Part 3: Reference
Bibliography
170 Holmes T.H. and Rahe R.H. (1967) The social
readjustments rating scales, Journal of Psychosomatic
Research, 11:213-218
181 HSE (1998) Manual handling. Manual Handling
Operations Regulations 1992. Guidance on
Regulations L23 (Second edition), HSE Books
190 HSE (2003) Organizational change and major
accident hazards, Chemical Information Sheet
CHIS7, HSE Books
171 Honey P. & Mumford A. (1992) The Manual
of Learning Styles. Peter Honey Publications,
Maidenhead
182 HSE (1998) Working with VDUs Leaflet
INDG36(rev1), HSE Books
www.hse.gov.uk/pubns/indg36.pdf (as of May 2008)
191 HSE, (2004) Work-related stress: A short guide, HSE
Books
172 HSE (no date) Successful Health and Safety
Management, HS(G)65, HSE Books
183 HSE (1999) Reducing error and influencing
behaviour HS(G)48 (Second edition) HSE Books
192 HSE (2004) International Stress Management
Association, ACAS, Working together to reduce
stress at work: A guide for employees
173 HSE Musculo-Skeletal Disorders
www.hse.gov.uk/msd/faq.htm (as of May 2008)
184 HSE (2000) Better Alarm Handling. Fact sheet www.
hse.gov.uk/pubns/chis6.pdf (as of May 2008)
193 HSE (2005) Tackling stress: The management
standards approach, HSE Books
174 HSE Manual Handling Assessment Charts (MAC),
www.hse.gov.uk/pubns/indg383.pdf (as of May
2008)
185 HSE (2000) Management of Health & Safety at
Work Regulations 1999 Approved Code of Practice
and Guidance (L21) (Second Edition), HSE Books
175 HSE stress website
www.hse.gov.uk/stress/index.htm (as of May 2008)
186 HSE (2000) Getting to grips with manual handling: A
short guide for employers, Leaflet INDG143(rev1)
HSE Books
194 HSE (2005), Development and validation of the
HMRI safety culture inspection toolkit, RR365
www.hse.gov.uk/research/rrhtm/rr365.htm (as of
May 2008)
176 HSE (no date) HSE Stress Standards
www.hse.gov.uk/stress/standards/index.htm (as of
May 2008)
177 HSE (1994) Manual handling: Solutions you can
handle. HS(G)115, HSE Books
178 HSE (1995) Improving Compliance with Safety
Procedures. HMSO, London
179 HSE (1997) Seating at work HS(G)57 (Second
edition), HSE Books
180 HSE (1997) Lighting at work. HS(G)38 (Second
edition) , HSE Books
Understanding Human Factors/June 08
187 HSE (2001) HSE Enforcement Statement/Quality
Statement for Continuing Aim 2: Document G Major incident response and investigation policy and
procedures
188 HSE (2003) Aching arms (or RSI) in small businesses
- Is ill health and sickness absence due to upper limb
disorders a problem in your workplace? HSE Books
Leaflet INDG171(rev1)
189 HSE (2003) Work with display screen equipment.
Health and Safety (Display Screen Equipment)
Regulations 1992. Guidance on Regulations (Second
edition) L26 HSE Books
195 HSE (in press) The development of a fatigue/risk
index for shiftworkers, to be published
196 Human Engineering Ltd. (2004) User Guide for
Human Factors SPAD Hazard Checklist: Issue 2
(December 2004 HEL/RSSB/041123/RTB02, RSSB
197 Human Engineering Ltd, (2004) The Human Factors
SPAD Hazard Checklist, Issue 2 HEL/RSSB/041123/
RTB02, Network Rail
198 Hyponen H. (1999) Focus Groups. In H. A. Williams
J. Bound & R. Coleman (eds) The Methods Lab: User
Research for Design. Design for Ageing Network
199 IEEE Guide for the application of human factors
engineering in the design of computer-based
Page 195
Part 3: Reference
Bibliography
monitoring and control displays for nuclear power
generating stations
200 IEEE Guide for the application of human factors
engineering to systems, equipment and facilities of
nuclear power generating stations
201 IEC 225 Octave-octave + 1/3 octave band filters
for the analysis of sound and vibrations, first edition.
International Electrotechnical Commission
202 IEC 845-04-57: (CIE 17.4) International Lighting
Vocabulary, International Electrotechnical
Commission
203 International Union of Railways, (2002) Layout
of driver’s cabs in locomotives, railcars, multipleunit trains and driving trailers, UIC leaflet 651, 4th
edition, July 2002
204 International Union of Railways (2004) SafeCulture:
A Method for Assessing Organisational Safety at
Interfaces
205 ISO/DIS 9335 Ergonomic principles for the design
of signals, 88/38160 DC
206 ISO 2575 (1995) BS AU 143f Road Vehicles,
Symbols for Controls, Indicators and Tell-tales
207 ISO 7000 (1989) Graphical Symbols for Use on
Equipment
208 ISO 9241-3 (1992) EN 29241-3 (1993) Ergonomic
Requirements for Office Work with Visual
Display Terminals (VDTs) Part 3 Visual Display
Requirements
Page 196
209 ISO 10075-1 (2000) Ergonomic principles related
to mental workload: General terms and definitions
219 Jankowicz D. (2003), The easy guide to repertory
grids, John Wiley
210 ISO 11428 (1996) Ergonomics – visual danger
signals – general requirements, design and testing
220 Jeannot E. (1999) Situation Awareness:
Recommendations to Project Leaders, EEC Draft
Note. Brétigny-sur-Orge, France: EUROCONTROL
Experimental Centre
211 ISO/IEC 10741-1 (1995) Information technology –
User system interfaces – Dialogue interaction Part 1
Cursor control for text editing
212 ISO/IEC 11581-1 (2000) Information technology –
User system interfaces and symbols – Icon symbols
and functions Part 1: Icons – General
213 ISO/IEC 11581-2 (2000) Information technology –
User system interfaces and symbols – Icon symbols
and functions Part 2: Object icons
214 ISO/IEC 11581-3 (2000) Information technology –
User system interfaces and symbols – Icon symbols
and functions Part 3: Pointer icons
215 ISO/IEC 11581-6 (1999) Information technology –
User system interfaces and symbols – Icon symbols
and functions Part 6: Action icons
216 ISO/TR 16982 (2002) Ergonomics of human-system
interaction – usability methods supporting humancentred design
217 ISO/TS 16071 (2003) Ergonomics of human-system
interaction – guidance of accessibility for humancomputer interface
218 Ison R. (2002) The design of ‘Learning Systems’:
experiences from the Open University, UK. ‘Towards
an Information Society for All 2 - New Pathways to
Knowledge’ conference, Berlin 8-9th March 2002
221 Jeannot E. Kelly C. & Thompson D. (2003) The
development of Situation Awareness measures
in ATM systems. EATMP report. HRS/HSP-005REP-01: © European Organisation for the Safety of
Air Navigation EUROCONTROL
222 Johnson P. Diaper D. & Long J. (1984) Tasks, skills
and knowledge: Task analysis for knowledge-based
descriptions. In B. Shackel (Ed.) Interact ‘84 - First
IFIP Conference on Human-Computer Interaction.
23-27, Elsevier, Amsterdam
223 Johnson, S. O., Herrera, I. A., Jersin, E., Rosness, R.,
Vatn, J., Veiseth, M., Tungland, M., Bergersen, C. E. B.
(2004) The Track to Safety Culture (SafeCulture)
– a toolkit for operability analysis of cross border
rail traffic, focusing on safety culture. SINTEF for
the UIC Herriot P. (1989) Recruitment in the 90’s,
Institute of Personnel Management
224 Kakabadse A. et al, (1988) Working in Organizations.
Penguin, especially Chapter 5
225 Kanki B.G. & Helmreich R.L. (1993) Cockpit
Resource Management, Academic Press
226 Kanter R.M. (1983) The Change Masters, Unwin
Paperbacks
Understanding Human Factors/June 08
Part 3: Reference
Bibliography
227 Katzenbach J.R. & Smith D.K. (1993) The Wisdom of
Teams, McGraw Hill
235 Kirwan B. (1994) A Guide to Practical Human
Reliability Assessment, Taylor & Francis
228 The Keil Centre Ltd (2003) Managing Safety Culture
in the UK Rail Industry: Report on the Review of
Safety Culture Tools and Methods, RSSB (available
from www.rssb.co.uk as of May 2008)
236 Kirwan B. (1996) The validation of three Human
Reliability Quantification techniques – THERP,
HEART and JHEDI: Part 1 – technique descriptions
and validation issues Applied Ergonomics, Vol 27, 6,
359 - 373
229 Keinan G. Meir E. & Gome-Nemirovsky T.
(1984) Measurements of risk takers’ personality.
Psychological Reports 55:163-167
230 Kennedy R. & Kirwan B. (1998) Development
of a Hazard and Operability-based method for
identifying safety management vulnerabilities in high
risk systems, Safety Science, Vol. 30, 249-274
231 Kirakowski J. (1996) The Software usability
measurement inventory: background and usage. In P.
Jordan B. Thomas & B. Weerdmeester (eds) Usability
Evaluation in Industry, Taylor & Francis
232 Kirwan B. (1990) Human reliability assessment. In J.
R. Wilson & E. N. Corlett (eds.), Evaluation of human
work: a practical ergonomics methodology (2nd ed.
pp. 921-968), Taylor & Francis
233 Kirwan B. (1992a) Human error identification in
human reliability assessment. Part 1: Overview of
approaches. Applied Ergonomics Vol. 23(5), 299
- 318
234 Kirwan B. (1992b) Human error identification
in human reliability assessment. Part 2: detailed
comparison of techniques. Applied Ergonomics, 23,
371-381
Understanding Human Factors/June 08
237 Kirwan B. (1997) The validation of three Human
Reliability Quantification techniques – THERP,
HEART and JHEDI: Part 2 – Results of validation
exercise Applied Ergonomics, Vol 28, 1, 17 - 25
238 Kirwan B. (1997) The validation of three Human
Reliability Quantification techniques – THERP,
HEART and JHEDI: Part 3 – Practical aspects of the
usage of the techniques Applied Ergonomics, Vol 28,
1, 27 - 39
239 Kirwan B. & Ainsworth L.K (1992) A Guide to Task
Analysis, Taylor & Francis
240 Kirwan B. Evans A. Donohoe L. Kilner A. Lamoureux
Atkinson T. & MacKendrick H. (1997) Human
Factors in the ATM System Design Life Cycle. FAA/
Eurocontrol ATM R&D Seminar, Paris, France.
http://atm-seminar-97.eurocontrol.fr/ (as of May
2008)
241 Kirwan B. & Kennedy R. (1998) Development
of a Hazard and Operability-based method for
identifying safety management vulnerabilities in high
risk systems, Safety Science, Vol. 30, Pages 249-274
Cognitive Task Analysis. Pp417-431. Lawrence
Erlbaum Associates
243 Klein G. & Armstrong A.A. (2003) Critical Decision
Method. In Stanton et al (Eds) Handbook of Human
Factors and Ergonomics methods, Taylor & Francis
244 Klein G.A. Calderwood R. & MacGregor D. (1989)
Critical Decision Method for Eliciting Knowledge.
IEEE Transactions on Systems, Man and Cybernetics,
19(3), 462-472
245 Kletz, T. (unpublished) An engineer’s view of human
error. Dept of Chemical Engineering, Loughborough
Uni. LE11 3TU
246 Klinger D.W. & Hahn B.B. (In Press) Team Decision
Requirement Exercise: Making Team Decision
Requirements Explicit. In N. Stanton
247 Kolb D. (1984) Experiential Learning, Prentice Hall
248 Kroemer K.H.E. & Grandjean E. (1997) Fitting the
Task to the Human: A Textbook of Occupational
Ergonomics (5th Ed), Taylor & Francis
249 Ladkin P. (2001) Causal System Analysis – Formal
Reasoning About Safety and Failure, University of
Bielefeld
250 Larson J.R. (1989) The dynamic interplay between
employees’ feedback-seeking strategies and
supervisors’ delivery of performance feedback,
Academy of Management Review, Vol 14, No 3
242 Klein G. (2000) Cognitive Task Analysis of Teams.
In J. M. Schraagen S. F. Chipman V. L. Shalin (Eds)
Page 197
Part 3: Reference
Bibliography
251 Lawton R. (1998) Not working to rule:
understanding procedural violations at work, Safety
Science Vol. 28, No.2, pp.77–95, 1998
257 Lim K.Y. & Long J. (1994) The MUSE Method for
Usability Engineering. Cambridge: Cambridge
University Press
252 Learning theories
http://tip.psychology.org/theories.html (as of May
2008) Summarises 50 or so different theories
of learning developed by a large number of
psychologists and researchers in the field of human
learning
258 London Chamber of Commerce and Industry:
24/7 Health effects: shift and night working – an
employer’s guide, Fact Sheet 10. July 2004
253 Leather, P. Pyrgas, M. Beale, D. & Lawrence, D. (1998)
Windows in the workplace: Sunlight, view and
occupational stress. Environment & Behavior, 30,
739-762
254 Leveson N. et al. (2004) Effectively Addressing
NASA’s Organizational and Safety Culture: Insights
from Systems Safety and Engineering Systems. Paper
was presented at the Engineering Systems Division
Symposium, MIT, Cambridge, MA March 29-31.
255 Lewis C. & Reiman J. (1993) Task centred user
interface design: A practical introduction. Boulder,
CO: University of Colorado. Shareware book
available from ftp.cs.colorado.edu/pub/cs/distribs/
clewis/HCI-Design-Book
256 Lewis C. Polson P. Wharton C. & Rieman J. (1990)
Testing a Walk-through Methodology for TheoryBased Design of Walk-Up-and-Use Interfaces. In
Proceedings of CHI’90 conference on Human
Factors in Computer Systems, 235-241, New York:
Association for Computer Machinery
Page 198
259 Loughborough University, Ergonomics Ltd & Aston
Business School, (2004) Recidivist risk takers who
work at height. HSE.
260 Lucas D. (2004) Moving closer to human factors
integration in the design of rail systems: A UK
regulatory perspective, Human Factors and
Ergonomics Society Conference, Delft, 27-29
October 2004
261 MacMillan J. Paley M.J. Entin E.B. & Entin E.E. (In
Press) Questionnaires for Distributed Assessment of
Team Mutual Awareness. In N.A. Stanton, A. Hedge.
K. Brookhuis. E. Salas. & H. Hendrick. (In Press) (eds)
Handbook of Human Factors methods, Taylor &
Francis
262 Management Standards Centre (2005)
www.management-standards.org (as of May 2008)
263 Maslow A. (1943) A Theory of Human Motivation,
Psychological Review, 50, 370-396
264 Maslow A. (1998) Maslow on Management, John
Wiley
265 Mason S. (1997) Procedural violations - causes,
costs and cures. In Human Factors in Safety-Critical
Systems, eds. Redmill F. and Rajan J. Butterworth
Heinemann
266 Maybury T. (Ed.) (1993) Intelligent Multi Media
Interfaces. AAAI Press/MIT Press
267 McGuffog A. Spencer M.B. Stone B.M. & Turner
C. (2005) Guidelines for the management and
reduction of fatigue in train drivers, for RSSB
268 McGuire, P. (2002) Rightsizing, not downsizing is key
to staffing success, American College of Physicians,
www.acponline.org/clinical_information/journals_
publications/acp_internist/may02/rightsize.htm
269 Meshkati N. (1991) Human Factors in Large-Scale
Technological Systems’ Accidents: Three Mile Island,
Bhopal, Chernobyl. Industrial Crisis Quarterly, 5,
131-154
270 Ministry of Defence (2006), MAP-01-011 Human
Factors Integration Technical Guide, Sea Systems
Group, TES-SSG-ShipDes, Defence Procurement
Agency, MOD Abbey Wood, Bristol, BS34 8JH
www.hfidtc.com/pdf/MAP-01-011.pdf (as of May
2008)
271 Militello L.G. & Militello J.B. (2000) Applied Cognitive
Task Analysis (ACTA): A practitioner’s toolkit for
understanding cognitive task demands. In J. Annett
& N. S Stanton (Eds) Task Analysis, 90-113, Taylor &
Francis
272 Militello L.G. & Robert J.B. Hutton Klein Associates
Inc. 582 E. Dayton-Yellow Springs Road, Fairborn,
Ohio 43524
Understanding Human Factors/June 08
Part 3: Reference
Bibliography
273 Mills A.J. & Murgatroyd S.J. (1991) Organizational
Rules: A Framework for Understanding
Organizational Action. Open University Press, Milton
Keynes
274 Morgan B.B. Salas E. & Glickman A.S. (1994) An
analysis of team evolution and maturation. The
Journal of General Psychology, 120(3), 277-291
275 Morris C.H. & Leung Y.K. (1999) An Empirical Study
of Auditory Warnings in Aircraft. Human-Computer
Interaction INTERACT ‘99, Volume I, Edited by M.A.
Sasse and C. Johnson. IOS Press, Amsterdam
282 NASA-STD-3000 (1995) Man-Systems Integration
Standards Volumes 1 and 2
283 Network Rail, DSM Post-SPAD Interview Process –
Interview Technique and Checklist
284 Newsham D.B. (1969) The Challenge of Change
to the Adult Trainee. Training Information. Paper 3.
London HMSO
285 Nielsen J. (1993) Usability Engineering, Morgan
Kaufmann
276 Morrison,T, (1993) Staff Supervision in Social Care.
An Action Learning Approach, Harlow, Essex,
England, Longman.
286 Nielsen J. & Molich R. (1990) Heuristic evaluation
of user interfaces. In J. C Chew & J. Whiteside
(eds), Empowering people: CHI 90 Conference
Proceedings (pp. 249 - 256) Monterey, CA: ACM
Press
277 Mullins L (1999) Management and Organisational
Behaviour, 5th Edition, Financial Times/Pitman
Publishing
287 Norman D. (1988) The Psychology of Everyday
Things, Doubleday
278 Multer J. Rudich R & Yearwood K. (1998) Human
Factors Guidelines for Locomotive Cabs,
U.S. Department of Transportation Federal
Railroad Administration, Office of Research and
Development, 400 7th Street, S.W. Washington, DC
20590
279 Mumford E. (1983) Designing Human Systems,
Manchester Business School
280 Mumford E. & Beekman G.J. (1994) Tools for change
and progress, CSG Publications
281 NASA Man-System Integration Standards
http://msis.jsc.nasa.gov/ (as of May 2008)
Understanding Human Factors/June 08
288 Norman D. & Draper S. (Eds) (1986) Usercentered system design: New perspectives on
human-computer interaction. Lawrence Earlbaum
Associates, Hillsdale, NJ
289 O’Hare D. Wiggins M. Williams A. & Wong W. (2000)
Cognitive task analyses for decision centred design
and training. In J. Annett & N. Stanton (eds) Task
Analysis, 170-190, Taylor & Francis
290 Office of Public Sector Information. The Railways
and Other Guided Transport Systems (Safety)
Regulations 2006 can be accessed from the Office
of Public Sector Information (OPSI) website at:
www.opsi.gov.uk/si/si2006/uksi_20060599_en.pdf
(as of May 2008)
291 Office of Rail Regulation, The Railways and Other
Guided Transport Systems (Safety) Regulations
2006 – Guidance on Regulations can be accessed
from the Office of Rail Regulation (ORR) website at:
www.rail-reg.gov.uk/server/show/ConWebDoc.7964
(as of May 2008)
292 Office of Rail Regulation (2007), Developing and
maintaining staff competence, 2nd Edition, ISBN
07176 1732 7
293 O’Neill. P. (2004) Developing a risk communication
model to encourage community safety from natural
hazards. Discussion Paper Jun 2004 for State
Emergency Services, Australia
294 Ormerod T.C. Richardson J. & Shepherd A. (1998)
Enhancing the usability of a task analysis method:
A notation and environment for requirements.
Ergonomics 41(11), 1642-1663. Reprinted in Annett
& Stanton (2000) op.cit. Pp. 114-135
295 Osborne D.J. (1987) Ergonomics at work. (2nd Ed)
Wiley-Interscience
296 Park L. (2004) Work-related deaths - Investigators
Guide, HSE. The Investigators Guide was produced
to assist those tasked with investigating deaths in the
workplace
297 Patrick J. (1992) Training Research & Practice,
Academic Press
Page 199
Part 3: Reference
Bibliography
298 Patrick J. Gregov A. & Halliday P. (2000) Analysing
and training task analysis. Instructional Science, 28(4),
51-79
308 Pheasant S. (2001) Bodyspace. Anthropometry,
Ergonomics and the Design of Work. 2nd Edition.
Taylor and Francis
317 Railway Safety, Engineering Safety Management
Yellow Book 3: Application Note 3 Human Error:
Causes, Consequences and Mitigations Issue 1.0
299 Patterson R.D. (1982) Guidelines for auditory
warning systems on civil aircraft. Civil Aviation
Authority Paper CAA 82017
309 Polson P.G. Lewis C. Rieman J. & Wharton C. (1992)
Cognitive walk-throughs: a method for theory based
evaluation of user interfaces. International Journal of
Man-Machine Studies, 36 pp741-773
318 RSSB, (2005) ‘Feeling tired’ resources
www.rssb.co.uk/expertise/human_factors/facts.asp
(as of May 2008)
300 Patterson R.D. (1990) Auditory Warning Sounds
in the Work Environment. In Human Factors in
Hazardous Situations, Edited by D.E. Broadbent, J.
Reason and A. Baddeley. Clarendon Press, Oxford
37-44
301 Pedler M. Burgoyne J. & Boydell T. (2001) A
Manager’s Guide to Self-development, McGraw-Hill
Education
302 Pell A. (2000) The Complete Idiot’s Guide to
Recruiting the Right Staff, Alpha Books
303 Pew R.W. Miller D.C. Feehrer C.S. (1981) Evaluation
of Proposed Control Room Improvements Through
Analysis of Critical Operator Decisions. EPRI-1982.
Electric Power Research Institute, Palo Alto, CA
304 Pheasant S. (1991) Ergonomics, Work and Health.
Macmillan
305 Pheasant S. (1996) Bodyspace: Anthropometry,
ergonomics and the design of work, Taylor & Francis
306 Pheasant S. PP 7317. Ergonomics Standards &
Guidelines for Designers, published by BSI
307 Pheasant S. (1991) Ergonomics, work and health,
Macmillan
Page 200
310 Preece J. et al (1994) Human-Computer-Interaction.
Addison Wesley
311 Preece J. Rogers Y. & Sharp H. (2002) Interaction
Design: Beyond human-computer interaction, John
Wiley
312 Quinones M.A. & Ehrenstein A. (1999), Training
for a Rapidly Changing Workplace, American
Psychological Association
313 Railways (Accident Investigation and Reporting)
Regulations (2005) Statutory Instrument 2005 No.
1992 www.opsi.gov.uk/si/si2005/20051992.htm#6
(as of May 2008)
319 RSSB (2005) Minimising the Impact of Railway
Suicides on Railway Staff, Research Project T317.
320 RSSB (2005) Safety Culture, Human Factors Fact
Sheet 1, Rail Safety
321 RSSB (2007) Good Practice in Training: a guide to
the analysis, design, delivery and management of
training, RS/220 Issue 2, June 2007
322 RSSB (2007) Good Practice Guide on simulation as
a tool for training and assessment, RS/501 Issue 2,
June 2007
323 RSSB (2008) Good Practice Guide on Assessment,
RS/701, Issue 2, Aug 2008, RSSB, Euston, London
314 Railway Group Standard GM/RT2161 (1995)
Requirements for Driving Cabs of Railway Vehicles,
Aug 1995
324 RSSB (2008) Good Practice Guide on Driver
Specific Assessment RS/702 Issue 1 Aug 2008, RSSB,
Euston, London
315 Railway Safety (1996), Railway (Safety Critical Work)
Regulations 1994. Approved Code of Practice, HSE
Books
325 RSSB (2008) Good Practice Guide on Cognitive
and Individual Risk Factors RS232, Rail Safety and
Standards Board
316 Railway Safety Principles and Guidance Part 3A,
Developing and Maintaining Staff Competence, HSE
Books
326 RSSB (2008) RIS-3751-TOM Rail Industry Standard
for Train Driver Selection Issue 1 Aug 2008, RSSB,
Euston, London
Understanding Human Factors/June 08
Part 3: Reference
Bibliography
327 Rasmussen J. Pejtersen A.M. & Goodstein L.P. (1994)
Cognitive systems engineering, John Wiley
328 Ravden S.J. & Johnson G.I. (1989) Evaluating Usability
of Human-Computer Interfaces: A practical method,
Chichester, Ellis Horwood
329 Rawlinson J.G. (1981) Creative thinking and
brainstorming, Gower, 1981
& N. Meshkati (Eds.), Human Mental Workload,
Elsevier, Amsterdam
337 Richards M. (1998) The Stress Pocketbook,
Management Pocketbooks
349 Schermerhorn J.R. Jr. (2005) Management, 8th
Edition, John Wiley
338 Robbins H. & Finley M (2000) Why Teams don’t
Work, Tenere
350 Schneiderman B. (2005) Designing the User
Interface: Strategies for Effective Human Computer
Interaction
330 Reason J. (1990) Human Error. New York:
Cambridge University Press
339 Robens Lord. (1972) Safety and Health at Work.
Report of the committee, 1970–1972. Open
University Press, Milton Keynes. HMSO, London
331 Reason J. (2000) Human Error: Models and
Management. British Medical Journal, 320: 768-770
340 Robertson I.S. (2001) Problem Solving, Psychology
Press
332 Reason J. (2001) Flight Safety Australia. Jan/Feb 2001
http://casa.gov.au/fsa/2001/jan/28-41.pdf (as of May
2008)
341 Roughton J. & Mercurio J. (2002) Developing an
Effective Safety Culture: A Leadership Approach,
Butterworth-Heinemann
333 Reason J. & Hobbs A. (2003) Managing maintenance
error, Ashgate Publishing
342 Roxburgh C. (2003) Hidden flaws in strategy,
McKinsey Quarterly, 2003, No. 2
334 Rebiffe P.R. (1969) Le Siege du Conducteur: son
Adaptation Aux Exigences Functionnelles et
Anthropometriques. Ergonomics, 1969. (From
McCormick and Sanders 1982, page 341)
343 Rubin Herbert J. & Rubin Irene S. (1995) Qualitative
interviewing: The art of hearing data. Thousand
Oaks, Ca.: Sage
335 Reed C. (2003) Recruiting the Right People, part of
the Human Resources Management Tools Project
by NetGain Partners Inc. for the Cultural Careers
Council and the Cultural Human Resources Council,
Ontario, Canada
336 Reid G.B. & Nygren T.E. (1988) The subjective
workload assessment technique: A scaling procedure
for measuring mental workload. In P. S. Hancock
Understanding Human Factors/June 08
348 Schein E.H. (1999) Corporate Culture – Survival
Guide, Jossey Bass
344 Safety Critical Communication Focus Group
www.rail-scc.co.uk (as of May 2008)
345 Salvendy G. (1997) Ed. Handbook of Human
Factors and Ergonomics 2nd Ed. John Wiley, Canada
346 Sanders M.S. & McCormick E.J. (1993) Human
factors in Engineering and Design (7th Ed) New
York: McGraw Hill International
347 Schein E. (1992) Organizational culture and
leadership, 2nd Edition, Jossey Bass
351 Scott, T, Mannion, R, Davies, & H, Marshall, M. (2003)
Healthcare performance and organisational culture,
Radcliffe Medical Press
352 Seely Brown J. & Duguid J. (1991) Organizational
Learning and Communities of Practice: Towards a
Unified View of Working, Learning, and Innovation,
The Institute of Management Sciences (now
INFORMS)
353 Selcon S.J. & Taylor R.M. (1989) Evaluation of the
Situation Awareness Rating Technique (SART) as
a tool for aircrew system design, Proceedings of
AGARD Symposium on Situational Awareness in
Aerospace operation, Copenhagen
354 Senge P., Kleiner A., Roberts C., Ross R., Roth G.,
& Smith B. (1999) The Dance of Change, Nicholas
Brealey
355 Shappell S.A. & Wiegmann D.A. (1997a) A human
error approach to accident investigation: The
taxonomy of unsafe operations. International J.
Aviation Psychology, 7(4), 269-291
356 Shappell S.A. & Wiegmann D.A. (1997b) A reliability
analysis of the Taxonomy of Unsafe Operations.
Page 201
Part 3: Reference
Bibliography
Aviation, Space, and Environmental Medicine, 68,
620
357 Shepherd A. (1985) Hierarchical Task Analysis
and training decisions, Programmed Learning and
Educational Technology, 22, 162-176
358 Shepherd A. (2002) Hierarchical Task Analysis, Taylor
& Francis
359 Silberman M (2003), The Consultants Big Book of
Organisational Development Tools’ McGraw-Hill
360 Situation Awareness Global Assessment Technique,
See SA Technologies www.satechnologies.com (as
of May 2008)
361 Stanton N.A. (2002) Human error identification in
human computer interaction. In: J. Jacko and A. Sears
(eds) The Human-Computer Interaction Handbook,
Lawrence Erlbaum Associates
365 Stanton N.A. & Young M.S. (1998) Is utility in the
mind of the beholder? A review of ergonomics
methods. Applied Ergonomics. 29 (1) 41-54
366 Stammers R & Patrick J. (1975) The Psychology of
Training. Essential Psychology, Ed P. Herriot, Methuen
367 Stevens R. Brook P. Jackson K. & Arnold S. (1998)
Systems engineering: coping with complexity,
Prentice Hall, Europe
368 Stone B.M. (2004) Evaluation of current tools and
techniques used for estimating risks associated with
shift patterns, QinetiQ Centre for Human Sciences
Report, QinetiQ/KI/CHS/CR032327, for RSSB
369 Stone D. Patton B. & Heen S. (1999) Difficult
Conversations, Penguin Books
370 SUMI website www.ucc.ie/hfrg/questionnaires/sumi/
index.html (as of May 2008)
362 Stanton N.A. & Baber C. (2002) Error by design:
methods to predict device usability. Design Studies,
23 (4), 363-384
371 Sutherland V. & Cooper C. (2000) Strategic Stress
Management: An Organizational Approach, Palgrave
MacMillan
363 Stanton, N. A., Salmon, P. M., Walker, G. H., Baber,
C., Jenkins, D. P. (2005). Human Factors Methods:
A Practical Guide for Engineering and Design.
Ashgate, Aldershot. ISBN 0-7546-4660-2 (HBk),
0-7546-4661-0 (PBk)
372 Swann C.D. & Preston M.L. (1995) Twenty-five
years of HAZOPs. Journal of Loss Prevention in the
Process Industries, Vol. 8, Issue 6, 1995, 349-353
364 Stanton N.A. & Wilson J. (2000) Human Factors:
step change improvements in effectiveness and
safety. Drilling Contractor, Jan/Feb, 46-41
Page 202
373 Swezey R.W. Owens J.M. Bergondy M.L. & Salas
E. (2000) Task and training requirements analysis
methodology (TTRAM): An analytic methodology
for identifying potential training uses of simulator
networks in teamwork-intensive task environments.
In J. Annett & N. Stanton (eds) Task Analysis, pp150 169, Taylor & Francis
374 Taylor R.M. (1990) Situational Awareness Rating
Technique (SART): The development of a tool for
aircrew systems design. In Situational Awareness in
Aerospace Operations (AGARD-CP-478) pp3/1
–3/17, Neuilly Sur Seine, France: NATO-AGARD
375 Teaming Up (2004) Components of Safety under
High Risk edited by Rainer Dietrich and Kateri
Jochum, published by Ashgate
376 TUC on stress - see www.tuc.org.uk/h_and_s/tuc10147-f0.cfm (as of May 2008)
377 Tuchman B.W. (1965) Developmental Sequence in
Small Groups, Psychological Bulletin, vol. 63, 1965,
pp. 384-399
378 Tufte E.R. (1990) Envisioning Information. Cheshire,
CT: Graphics Press
379 Tullis T.S. (1990) The Formatting of Alphanumeric
Displays: A Review and Analysis in Venturino M.
Ed. Selected Readings in Human Factors. Human
Factors Society (1990)
380 Usability evaluation www.usabilityhome.com/ (as of
May 2008). This website summarises several usability
evaluation methods organised under the three types
of: Testing, Inspection, and Inquiry
381 Usability testing - see http://jthom.best.vwh.net/
usability/usable.htm (as of May 2008)
382 Vanderhaegen F. (2001) A non-probabilistic
prospective and retrospective human reliability
analysis method - application to railway system,
Understanding Human Factors/June 08
Part 3: Reference
Bibliography
Reliability Engineering and System Safety 71 (2001)
1–13
383 Van der Schaaf T.J. & Kanse L. (2002) Not reporting
successful recoveries from self-made errors? An
empirical study in the chemical process industry.
In Workshop on the Investigation and Reporting
of Accidents 2002, (IRIA 2002) Ed C. W. Johnson.
GIST Technical Report F2002-2, Department Of
Computing Science, University Of Glasgow, Scotland
384 Van Gundy A.B. (1981) Techniques of structured
problem solving, van Nostrand- Reinhold, New York
385 Van Welie M. & Van Der Veer G. (2003) Groupware
Task Analysis. In E. Hollnagel (Ed) Handbook of
Cognitive Task Design. 447 - 477. Lawrence Erlbaum
Associates
386 Verplank B. Fulton J. Black A. & Moggridge B. (1993)
Observation and Invention – Use of scenarios in
interaction design. Tutorial notes for InterCHI 93.
Amsterdam
387 Vicente K.J. (1999) Cognitive Work Analysis: Toward
Safe, Productive, and Healthy Computer-Based
Work. Lawrence Erlbaum Associates
388 Vidulich M.A. & Tsang P.S. (1985) Assessing
subjective workload assessment. A comparison of
SWAT and the NASA bipolar methods. Proceedings
of the Human Factors Society 29th Annual Meeting.
Santa Monica, CA: Human Factors Society, pp 71-75
389 Vidulich M.A. & Tsang P.S. (1986) Collecting NASA
Workload Ratings. Moffett Field, CA. NASA Ames
Research Center
Understanding Human Factors/June 08
390 Vidulich M.A. & Tsang P.S. (1986) Technique of
subjective workload assessment: A comparison of
SWAT and the NASA bipolar method. Ergonomics,
29 (11), 1385-1398
391 Watts L.A. & Monk A.F. (2000) Reasoning about
tasks, activities and technology to support
collaboration. In J. Annett & N. Stanton (Eds) Task
Analysis, Taylor & Francis
392 Wedderburn A. (ed) (2000) Shift work and Health,
Bulletin of European Studies on Time (BEST
1/2000), European Foundation for the Improvement
of Living and Working Conditions
www.eurofound.europa.eu/publications/htmlfiles/
ef0009.htm (as of May 2008)
393 Weick K. & Sutcliffe K. (2001) Managing the
Unexpected: Assuring High Performance in an Age
of Complexity. Jossey Bass Wiley
394 Weinger M.B. (2000) Auditory Warnings in the
Medical Work Domain: An Overview of Critical
Issues. Proceedings of the XIVth Triennial Congress
of the International Ergonomics Association and
44th Annual Meeting of the Human Factors and
Ergonomics Society, San Diego, California, USA, July
29-August 4, 2000. Human Factors and Ergonomics
Society, Santa Monica, California, USA, Volume 1
395 West M. (1994) Effective Teamwork, British
Psychological Society
396 Whalley (1988) Minimising the cause of human
error. In B. Kirwan & L. K. Ainsworth (eds.) A Guide
to Task Analysis. Taylor & Francis
397 Wharton C. Rieman J. Lewis C. & Polson P.
(1994) The Cognitive Walk-through Method: A
Practitioners Guide, John Wiley
398 Wickens C.D. (1992) Engineering Psychology and
Human Performance. 2nd Ed. Harper Collins
399 Wilde G.S.J. (1998) Risk homeostasis theory: an
overview. Injury Prevention 4, 89-91 British Medical
Journal Publishing Group
400 Wilkins P.A. (1981) Assessing the effectiveness of
auditory warnings. British Journal of Audiology 15
263-274
401 Williams J.C. (1986) HEART – a proposed method
for assessing and reducing human error. In 9th
Advances in Reliability Technology Symposium,
University of Bradford
402 Williams A. & Cooper L. (2002) Managing
Workplace Stress: A Best Practice Blueprint, John
Wiley
403 Wilson J. & Rajan J. (1995) Human-machine
interfaces for systems control. In Evaluation of
Human Work: A Practical Ergonomics Methodology,
Eds J.R. Wilson and E.N. Corlett, Taylor & Francis
404 Wilson J.R. & Corlett E.N. (Eds) (1995) Evaluation of
Human Work: A Practical Ergonomics Methodology,
Eds, Taylor & Francis
405 Woods D.D. (1991) The cognitive engineering of
problem representations. In G.R.S. Weir & J.L. Alty
(Eds), Human-computer interaction and complex
systems, Academic Press
Page 203
406 Woodson W.E. Tillman B. & Tillman P. (1992) Human
Factors Design Handbook. 2nd Edition. McGraw-Hill
407 Zeitlin L.R. (1994) Failure to follow safety
instructions: faulty communication or risky decisions?
Human Factors 36, 172–181
understanding
Index
human factors
understanding
Index
human factors
Part 3: Reference
Index
Index
A
Accident
causation 19
HF checklist 20
interviewing 20
investigation 20
near-miss 21
Alarms 32, 33, 39, 44
Appraiser bias 70, 71
Armagh train disaster 9
Arousal 121
Attention 11, 29, 39, 43
Automation
ironies 38
strategy 38
B
Behaviour modification 17
C
Change
culture 112
drivers and obstacles 109
management principles 113
organisational 111
technology-based 110
Colour
psychological effects 43
Communication
and safety culture 107
Understanding Human Factors/June 08
failures 106
good practice 108
importance 106
improving 108
managing 108
Competence assessment 69
Confidential reporting 22
Controls 32
Culture
assessment 94
definition 92
organisational 88, 91, 92, 110, 112
safety 93
D
Design
cab 25, 36, 59
controls 32
equipment/interface 31, 33, 41, 127
job 49
workplace 19, 41, 42, 46, 48, 127
user-centred
approach 27
importance 25
requirements 25
Difficult conversation
definition 73
E
Equipment design
principles 31
Errors
causes 10
reducing 11
types 9
F
Fatigue 128
Fitt’s List 37
Function allocation
definition 37
dynamic 40, 181
fixed 37, 38, 40
rules 38
strategies 37
H
High reliability organisation 4, 94, 95, 101
Human Factors
definition 3
five areas 3
payoff 4
Human Factors Integration Plan 27, 28, 181
I
Indoor climate 45
Interface design 35, 36
Page 207
Part 3: Reference
Index
principles 31
railway specific 36
Interviewing
after an accident 20
biases 70
competency 80
selection 80, 82
subject matter experts 56
Investigator biases 21
J
Job design
example 51
principles 50
L
Leadership
and safety culture 97
definition 96
good qualities 98
style 96
Learning
effectively 61
Lighting 43, 46
Logical reasoning 39
Longford explosion 67
M
P
Maintainers 29
Management
and safety culture 101
definition 99
Manual workers 45
Mental capacities 39
Mistakes
causes 10
reducing 11
types 9
Morale
definition 117
improving 119
Motivation
definition 117
Herzberg 118, 119
improving 119
Maslow 118
psychological contract 119
Perception 16, 20, 34, 36, 39, 46
Performance
appraisal 68, 69, 70
knowledge-based 9, 10
rule-based 9
skill-based 9
Posture
requirements 42
N
Near-miss 21, 22
Negotiation 74
Noise 35, 44
O
R
Recruitment
attracting interest 85
methods 84
right sizing 86
References (applicant) 83
Retention
approaches 87
importance 87
Risk
comfort zone 13, 15
perception 16
taking 11, 15, 16
Rules
breaking 12, 13
credibility 14
reducing rule-breaking 13
Organisation
culture 91
right sizing 86
safety culture 93
Page 208
Understanding Human Factors/June 08
Part 3: Reference
Index
S
T
V
Safety culture 93
Sedentary workers 45
Selection
assessment centres 81
biodata 80, 81
discrimination 82
interview 80
methods 80, 81
process 79
psychometric tests 80, 81
references 83
Shift work
problem management 129, 130
problems 128
Simulation 14, 34, 42, 50, 58, 64, 69, 71, 127
Situation awareness
definition 34
design challenge 34
measurement 34
Skill fade 64
SPAD 19, 20, 22, 102
Hazard Checklist 20, 22
Stakeholder analysis 25, 26, 27
Stress
causes 121
definition 120
effects 121
recognising 123
reducing 123
Style guide 33
Swiss cheese model 18
Systems engineering 27, 28
Task analysis
cognitive 49
definition 47
hierarchical 47
purpose 48
Team development 71, 72
Teamworking
definition 103
good practice 104
improving 105
Techniques
summary 135
Training
computer-based 58
design 62
embedded 58, 193
evaluation 65, 66, 67
learner-centred 64
media options 58
media selection 57
minimalist approach 62, 63
needs analysis 55
part-task 62, 64, 67
skill fade 64
skills 62
Vibration 44
Vlad the Impaler 17
Understanding Human Factors/June 08
W
Workload
definition 125
identifying problems 127
measurement 126
Workplace design
cab design 46
colour 43
indoor working 45
lighting 43
noise 44
principles 41
shape, proportion 43
vibration 44
U
Usability
testing 28, 29
User-centred design
approach 27
importance 25
requirements 25
Page 209
Understanding Human Factors – a guide for the railway industry
June 2008
Photographs and technical materials reproduced with permission
© 2008 Rail Safety and Standards Board
ISBN 0-9551435-3-5
ISBN 978-0-9551435-3-3
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement