User manual | Performance Measures for Snow and Ice Control Operations

NCHRP
Web-Only Document 136:
Performance Measures for Snow
and Ice Control Operations
T.H. Maze
Chris Albrecht
Dennis Kroeger
And
Jon Wiegand
Center for Transportation Research and Education
Iowa State University
Ames, Iowa
Contractor’s Final Report for NCHRP Project 6-17
Submitted December 2007
National Cooperative Highway Research Program
ACKNOWLEDGMENT
This work was sponsored by the American Association of State
Highway and Transportation Officials (AASHTO), in cooperation with
the Federal Highway Administration, and was conducted in the
National Cooperative Highway Research Program (NCHRP), which is
administered by the Transportation Research Board (TRB) of the
National Academies.
COPYRIGHT PERMISSION
Authors herein are responsible for the authenticity of their materials
and for obtaining written permissions from publishers or persons who
own the copyright to any previously published or copyrighted material
used herein.
Cooperative Research Programs (CRP) grants permission to
reproduce material in this publication for classroom and not-for-profit
purposes. Permission is given with the understanding that none of the
material will be used to imply TRB, AASHTO, FAA, FHWA,
FMCSA, FTA, Transit Development Corporation, or AOC
endorsement of a particular product, method, or practice. It is expected
that those reproducing the material in this document for educational
and not-for-profit uses will give appropriate acknowledgment of the
source of any reprinted or reproduced material. For other uses of the
material, request permission from CRP.
DISCLAIMER
The opinion and conclusions expressed or implied in the report are
those of the research agency. They are not necessarily those of the
TRB, the National Research Council, AASHTO, or the U.S.
Government.
This report has not been edited by TRB.
CONTENTS
LIST OF FIGURES ...................................................................................................................... III
LIST OF TABLES........................................................................................................................ IV
ACKNOWLEDGMENTS ..............................................................................................................V
ABSTRACT................................................................................................................................. VII
SUMMARY.....................................................................................................................................1
CHAPTER 1. INTRODUCTION TO NCHRP 06-17 .....................................................................2
1.1. Performance Measurement ...........................................................................................2
1.2. Measurement of Winter Maintenance Performance .....................................................3
1.3. Methods for Measuring Winter Maintenance Performance .........................................4
1.4. Putting Winter Maintenance Performance Measurement into Context ........................5
1.5. Summary of Synthesis Findings and Assessment of Performance Measures...............5
1.6. Conclusions...................................................................................................................7
CHAPTER 2. LITERATURE REVIEW .........................................................................................8
2.1. Summary of Key Points ................................................................................................8
2.2. Background ...................................................................................................................9
2.3. Types of Performance Measures.................................................................................11
2.4. Performance-Based Levels of Service ........................................................................17
2.5. Performance-Based Pay Items in Contracts................................................................17
2.6. Innovative Use of Technology for Performance Measurement..................................18
2.7. Winter Weather Severity Indexes ...............................................................................18
CHAPTER 3. WINTER MAINTENANCE MEASURES USED BY HIGHWAY AGENCIES .24
3.1. Survey Results ............................................................................................................24
3.2. Summary and Conclusions .........................................................................................43
CHAPTER 4. SELECTED PRACTICES FOR PERFORMANCE MEASUREMENT ...............44
4.1. State Transportation Agencies in the United States ...................................................44
4.2. Summary .....................................................................................................................63
CHAPTER 5. SYNTHESIS AND ASSESSMENT OF PERFORMANCE MEASURES............63
5.1. Introduction.................................................................................................................63
5.2. Performance Measures in Use ....................................................................................64
5.3. Screening of Approaches ............................................................................................70
5.4. Summary of Approaches.............................................................................................75
CHAPTER 6. CREATING PERFORMANCE MEASURES TOOLBOX FOR SNOW AND ICE
CONTROL OPERATIONS...............................................................................................76
B-1
6.1. Introduction.................................................................................................................76
6.2. Benefits of Using Performance Measurement ............................................................76
6.3. Conclusions.................................................................................................................86
CHAPTER 7. DEVELOPING A FIELD TEST PLAN.................................................................87
7.1. Measures of Input, Output, and Operational Efficiency.............................................88
7.2. Measures of Highway Safety ......................................................................................89
7.3. Measures of Highway Mobility ..................................................................................90
7.4. Measures of Public Satisfaction..................................................................................91
7.5. Measures of Environmental Impacts ..........................................................................91
7.6. Conclusions.................................................................................................................93
REFERENCES ..............................................................................................................................95
ABBREVIATIONS .......................................................................................................................99
APPENDIX A. WINTER MAINTENANCE OPERATIONS PERSONNEL SURVEY
INSTRUMENT................................................................................................................A-1
APPENDIX B. RESPONSES TO SURVEY...............................................................................B-1
APPENDIX C. AGENCIES USING PUBLIC SATISFACTION SURVEYS ..........................C-1
ii
LIST OF FIGURES
Figure 1. Relationship between inputs, outputs, and outcomes.......................................................5
Figure 2. Personnel used for snow and ice control ........................................................................26
Figure 3. Evaluation of performance .............................................................................................27
Figure 4. Performance and payment ..............................................................................................28
Figure 5. Performance measures used by various agencies...........................................................29
Figure 6. The frequency with which targets are set.......................................................................31
Figure 7. Data sources for performance measurement ..................................................................33
Figure 8. Agencies using public satisfaction surveys ....................................................................34
Figure 9. Performance measures and road characteristics.............................................................35
Figure 10. Performance objectives and storm characteristics........................................................36
Figure 11. Agencies using performance measurement used for non-storm events .......................37
Figure 12. Agencies using storm severity index used ...................................................................38
Figure 13. Methods to report road condition .................................................................................39
Figure 14. Agencies segmenting highway areas............................................................................40
Figure 15. Agencies identifying benefits.......................................................................................41
Figure 16. Map of Minnesota districts and regain time.................................................................89
iii
LIST OF TABLES
Table 1. Sample PSIC table ...........................................................................................................14
Table 2. Locations responding to the survey .................................................................................25
Table 3. Performance measures and performance levels...............................................................32
Table 4. Minnesota DOT Pavement Regain Time by Roadway Class ..........................................50
Table 5. Finnish Road Network (Finnish Road Maintenance 2001) .............................................56
Table 6. Quality standards and friction indicators (Finnish Road Maintenance 2001) .................57
Table 7. Friction indicators and driving conditions (Finnish Road Maintenance 2001) ...............57
Table 8. Quality standards for snow removal (Finnish Road Maintenance 2001) ........................58
Table 9. Japanese road management performance plan.................................................................61
Table 10. Snow and ice operations for Sapporo, Japan .................................................................62
Table 11. Winter road standards and LOS for Sapporo, Japan (PIARC 2006) .............................63
Table 12. Summary of snow and ice control performance measures by category ........................65
Table 13. Agencies using various performance measures .............................................................66
Table 14. Outcome measures and approaches used by responding agencies ................................69
Table 15. Linking program activities and outputs to strategic objectives .....................................78
Table 16. Identifying the key program activities and outputs .......................................................79
Table 17. Identifying key snow and ice control issues and affected stakeholder groups ..............80
Table 18. Defining results..............................................................................................................81
Table 19. Performance requirements relative to responses and results .........................................82
Table 20. Establishing potential performance measures ...............................................................83
Table 21. Establishing baselines for measures ..............................................................................83
Table 22. Quality criteria for performance measures ....................................................................84
Table 23. Mn/DOT ranges for bare pavement regain time............................................................85
Table 24. A screening tool for quality considerations ...................................................................85
Table 25. Establishing accountability for implementation ............................................................86
Table 26. Identifying resource requirements for implementation .................................................86
iv
ACKNOWLEDGMENTS
T.H. Maze is a professor of Civil, Construction, and Environmental Engineering at Iowa State
University.
Chris Albrecht is an Assistant Scientist at the Center for Transportation Research and Education
at Iowa State University.
Dennis Kroeger is an Assistant Scientist at the Center for Transportation Research and Education
at Iowa State University.
Jonathan Wiegand is a graduate research assistant in the Department of Civil, Construction, and
Environmental Engineering at Iowa State University.
The authors would like to thank Wilfrid Nixon, professor of Civil and Environmental
Engineering at the University of Iowa, for his contributions to this project, especially in the areas
of snow and ice control procedures and winter weather indices.
The authors would also like to thank Stephen Andrle, former director of the Center for
Transportation Research and Education at Iowa State University, for his contributions to this
project.
v
ABSTRACT
Under the NCHRP 06-17 project, the research team surveyed snow and ice control organizations
in the United States, Canada, Europe, and Asia to determine the current trends in performance
measurement. The team also inquired about the methods used in developing these programs in
order to determine a practical, user friendly method to assist snow and ice control managers in
developing a performance measurement system that uses traditional and nontraditional
performance indicators and measurement issues. To achieve the project objectives, the
researchers issued a survey to snow and ice control agencies throughout North America, Europe,
and Asia to obtain data of the performance indicators and measures used, if any, by these
agencies. The identified performance indicators and measures were then categorized, defined,
and assessed for their usefulness. A process was then developed to assist snow and ice control
operations managers in preparing a customer-focused, environmentally friendly performance
measurement program.
vii
SUMMARY
The issue of performance measurement for snow and ice control has been a topic of much
interest. Developing meaningful data for snow and ice control has produced a variety of
responses and differing goals and objectives. However, a rigorous process that the snow and ice
control industry can use to determine the most appropriate performance measures and indicators
has been lacking.
Research was needed to examine current trends and issues and develop a process that can be
used by snow and ice control agencies to prepare a performance measurement system that is
sensitive to organizational and public needs as well as environmental concerns. This process
would provide a context, or framework, to select and apply appropriate performance indicators
and measures that are integral to snow and ice control decision-making. The research would also
analyze the different dimensions along which an agency’s performance could be defined,
measured, and interpreted based on an agency’s goals and objectives.
Under the NCHRP 6-17 project, the research team surveyed snow and ice control organization in
the United States, Canada, Europe, and Asia, to determine the current trends in performance
measurement. The team also inquired about the methods used in developing these programs in
order to determine a practical, user-friendly method to assist snow and ice control managers in
developing a performance measurement system that uses traditional and nontraditional
performance indicators. The plan provides a list of options of performance indicators and
measures, and explains how to incorporate the indicators and measures in the decision making
process to monitor and improve snow and ice control operations.
One method to incorporate the use of performance measures for snow and ice control operations
is a “toolbox” developed by the research team. This toolbox was designed for managers to use to
evaluate relevant performance measures for snow and ice control operations and assist them in
their decision making process.
To achieve the project objectives, the researchers first reviewed pertinent literature and research
findings in the area of performance measurement systems. Next, a survey was issued to snow
and ice control agencies throughout North America, Asia, and Europe to obtain data of the
performance indicators and measures used, if any, by these agencies. These performance
indicators and measures were then categorized by functional type and were fully defined. An
assessment of the usefulness of each was prepared. The research team then summarized the
theory and practice of the performance measurement. The performance measures were then
identified by their key aspects and identifying performance indicators and measures that may
have applicability in snow and ice control operations. A process was then developed to assist
snow and ice control operations managers in preparing a customer focused, environmental
friendly performance measurement program.
1
CHAPTER 1. INTRODUCTION TO NCHRP 06-17
The purpose of this research is to identify and assess the measures used to evaluate the
performance of winter maintenance activities (snow and ice removal from roadways) and to
recommend the most promising measures for further development. The research was conducted
in two parts. The first part entailed a comprehensive review of performance measures that have
been and are currently being used by transportation agencies. This work was accomplished
through a thorough review of the literature and a survey of dozens of agencies with winter
maintenance responsibilities. In the second part, the performance measures that offered the most
promise were identified. In other words, these were measures with the most potential to be
applied economically to a roadway network and provide reliable, repeatable, and comparable
measures of performance. These most promising measures were then recommended for further
development and use by highway agencies.
1.1. Performance Measurement
For many transportation agencies, performance measurement has become a critical issue in the
last five to ten years. Such that transportation agencies often attempt to tie strategic direction and
agency mission to performance measures. As Osborne and Gaebler, 1992, state in their popular
book Reinventing Government, “If you don’t measure results, you can’t tell success from failure.
If you can’t see success, you can’t reward it. If you can’t reward success you are probably
rewarding failure. If you can’t see success, you can’t learn from it. If you can’t recognize failure,
you can’t correct it. If you can demonstrate results, you can win public support.”
Performance measurement is one component of a larger “quality in government services”
movement. The growing emphasis on performance measurement by transportation agencies has
not been sufficiently considered because there was no such need to measure performance but
also due to two factors:
1. Transportation agencies have historically focused on standards and specification for
physical conditions or level of service (LOS). Generally, transportation agencies have
defined the LOS or conditions of a facility based on static standards. Only recently,
through asset management application, have these agencies begun to treat LOS and
conditions as a variable against which other financial and condition considerations can be
balanced.
2. The recent expansion of information technology and the ability to collect information that
would have been too costly or impossible to collect in the past has made the collection of
performance-related data possible. In addition, the public and public policy makers’
expectation for performance information has grown as they become accustom to having
information at their fingertips. Thus, the growing ability to provide more performance
information has driven the demand for collecting and reporting more performance
information.
2
Winter maintenance of roadways is a core and critical business element of many state
transportation agencies, and because it is a core business, it needs to be managed. For such
management, the performance of winter maintenance must be measured.
1.2. Measurement of Winter Maintenance Performance
Although winter maintenance is a critical activity, there are no standard methods for measuring
performance for either agency programs or those performed by contractors. The lack of standard
measures also makes it difficult to effectively manage and control winter maintenance activities
and subsequently impossible to benchmark and make comparisons both between and within
maintenance programs. Measuring the performance of winter maintenance makes it possible to
make intelligent management trade-offs between agency costs and user costs.
The agency costs of winter maintenance are quite significant. The direct costs have been
estimated to be at least $1.5 billion per year in the United States alone (NCHRP 5-26). On the
user cost side, it is difficult to determine the safety and mobility problems as a result of either not
performing winter maintenance or not performing winter maintenance effectively. However, it
has been shown that during snow storms of 0.2 inches of snowfall per hour or more that crash
rates (crashes per million vehicle miles) on the Iowa rural freeway system increase by 13 times
and increase even more during severe storm (low visibility and high winds) events. Failure to
remove snow and ice would only continue to extend these high crash rates beyond the end the
storm. Through performance measurement, a winter maintenance manager can control and direct
winter maintenance to make the best use of available resources and to reduce potential user costs
of travel.
Agencies currently measure winter maintenance performance from one or more of three basic
perspectives:
•
Inputs. Input measures represent the resources spent or utilized to perform snow and
ice control operations. These resources include fuel usage, labor hours, machinery or
equipment hours, and units of anti-icing materials or abrasives. The level of inputs is
directly proportional to agency costs and, therefore, they most easily and most
commonly are measured by transportation agencies. Because inputs are applied at the
beginning of the winter maintenance process, they are unable to help management
assess the efficiency, quality, and effectiveness of winter maintenance.
•
Outputs. Outputs quantify the resulting physical accomplishment of work performed
using resources in winter maintenance. Outputs might include the lane-miles plowed
or sanded, the number of lane-miles to which deicing materials were applied, lanemiles of anti-icing brine applied, and other accomplishments of the maintenance
process in units of work. Outputs are generally more useful than inputs alone because
inputs and output together can help to define the efficiency of winter maintenance
operations by determining what level of input was or will be required to achieve a
level of output. These measures may also be based on time and storm event.
3
•
Outcomes. Outcomes generally attempt to assess the effectiveness of the winter
maintenance activity, very often from the perspective of the user or customer.
Outcomes are inherently more difficult to measure. A desired outcome of winter
maintenance might include the improvement of safety, mobility, and/or user
satisfaction. Safety, mobility, and user satisfaction are abstract concepts and,
therefore, are measured through indicators known to be related to the desired
outcome. For example, safety might be measured through pavement friction or
through the reduction in number of crashes.
Other known outcome measures include bare pavement regain time, duration and
frequency of closures, advanced warning time to customers, and customer satisfaction
indicated by customer satisfaction surveys. Although conceptually it may appear
simple to measure outcomes, the measurement methods are generally complex. For
example, while the number of crashes during and following a storm can be quickly
(within a week) identified through a centralized crash record data base, crashes alone
do not indicate the relative risk of having a crash. Crash risk is measure by crashes
per vehicle miles traveled (VMT), a measure of exposure. Estimating VMT during
and immediately following a winter storm may be possible for managed urban
freeway systems but difficult to measure reliably for rural network of highways.
1.3. Methods for Measuring Winter Maintenance Performance
Friction has become a very attractive performance measure for snow and ice removal. Sweden
and Finland have been measuring friction for over 10 years. Japan also correlates friction with
crashes and traffic speed and volumes (PIARC 2005). However, there are different methods for
measuring friction. For example, several different types of friction measuring devices can be
mounted under winter maintenance trucks or towed by a supervisor’s vehicle. Once a technology
has been selected, decisions have to be made regarding the number of friction reporting devices
and the frequency of measures required to understand the snow and ice control performance
across a network of roadways.
In the U.S., a common measure for performance of winter maintenance is time to bare pavement.
The Minnesota Department of Transportation (Mn/DOT), for example, measures the time to
reach bare pavement throughout the state’s trunk highway system and has set different levels of
satisfactory performance depending on the level of traffic on the route. These levels were set
based on significant stakeholder input. In Minnesota, more heavily traveled routes have shorter
time to bare pavement goals. Underlying Mn/DOT’s performance measurement are standards for
identifying when the pavement is bare, data collection and entry techniques, and quality
assurance methodologies.
In general, to allow comparisons across jurisdictions or between jurisdictions, a common
reliable, and repeatable performance measure must be identified together with a specific
methodology for collecting compiling the relevant data, over the same time frames, and made
comparable by normalizing their relative severity (e.g., it is meaningless to compare a
performance when a blizzard takes placed in one jurisdiction while another only experiences
light snow.).
4
1.4. Putting Winter Maintenance Performance Measurement into Context
When making comparisons between and among jurisdictions, differences in the severity of
storms must also be taken into account, because the severity of a storm impacts the performance
of winter maintenance. Figure 1 illustrates the relationship between inputs, outputs, outcomes,
and the environment. As shown on the top of the figure, some of the environmental inputs.
Labor, equipment, and materials inputs for removing snow and ice from the roadway network,
are shown on the bottom. The results achieved from these inputs under these environmental
conditions are also shown in the figure.
Terrain & Solar
Geography Energy
Environmental
Conditions =
Storm Severity
Preci
Change in
Tem
Win
Spee
Tem
Desired Outcome =
Customer
Satisfaction
Remove Snow and Ice - Outputs
Inputs = Labor,
Equipment,
Materials,
Management,
and Information
quality and
quantity
Number and Abrasives Sal
Anti Cycl
Icing Length Type of Truck
RWI
Operations
management
Time to Bare
Pavement
Figure 1. Relationship between inputs, outputs, and outcomes
In this case, satisfying the customer (the road users) was chosen as the desired outcome, and
because shorter time to bare pavement reflects higher levels of satisfaction, time to bare
pavement is the resulting performance measure. The measurement of time to bare pavement must
be supported by a specific data collection methodology.
1.5. Summary of Synthesis Findings and Assessment of Performance Measures
Although a significant amount of published materials deals with different types of performance
measures, both in use and theoretical, a limited amount of literature documenting agencies’
utilization of performance measures in day-to-day practice. Various instances of research and
testing of proposed performance measures were described in literature, but often without
implementation or field testing. It appeared that some European countries and Japan are more
progressive in terms of developing and implementing winter maintenance performance
measures, likely because more snow and ice control operations are contracted to private
companies internationally than in the United States.
The survey of winter maintenance personnel was sent to 162 agencies covering the U.S. Snow
Belt states, Canadian provinces, northern Europe, and Japan. In all, 43 agencies responded. The
responses included agencies that did no snow and ice control performance measurement to those
5
that incorporated performance measures into their management plans. Most performance
measures cited by the respondents are tied to their accounting and management systems. These
measures include lane-miles plowed, personnel and/or overtime hours, tons of material used,
amounts of equipment deployed, and cost of operations.
Other measures used by some of the respondents include time to bare pavement, time to return to
a reasonably near-normal condition, length of road closures, and customer satisfaction. The
majority of the measures critical to the respondents’ snow and ice control operations focused on
public safety and mobility. Obviously, these subjects are central to the role of all transportation
agencies, so it makes sense that the measures would focus on them. By maintaining mobility and
traffic flow, accidents are reduced and public safety is enhanced.
The survey also found that, while state and local agencies are generally interested in providing
the best service to the public, budget and staffing constraints make it difficult for agencies to
experiment with new methods or technologies. For example, measures such as friction were
identified by only a few agencies and are generally farther from full-scale implementation,
especially in the United States.
The survey analysis identified four input measures, five output measures, and 11 outcome
measures used by public agencies to measure snow and ice control performance. A complete list
of the performance measures identified is provided in Chapter 5. To identify measures and
approaches that warrant further study, the following criteria were applied to the measures and
approaches:
Measure Criteria
•
•
•
•
•
Does the measure directly measure safety, mobility, or public satisfaction?
Does the measure improve snow and ice control?
Is the measure mapped to roadway segments?
Is the measure reported for garages or districts?
Is the measure sensitive to storm characteristics?
Approach Criteria
•
•
•
•
•
Is the approach quantitative?
Is the approach stable across observers?
Is the technology likely to improve?
Is a major capital or operational investment required?
Can the approach be piggybacked on another system to reduce installation costs?
The assessment presented in Chapter 5, determined that outcome measures should be pursued
further, the measurement of snow and ice control is to have a role in improving safety and
mobility. To help determine the measures and approaches to pursue further, the 11 outcome
measures identified in this study were reduced to three basic categories, and two approaches
were identified for each measure.
6
Measure: Degree of clear pavement
•
•
Approach: Manual observation
Approach: Camera-assisted observation
Measure: Traffic flow
•
•
Approach: Detectors – speed, volume, and occupancy
Approach: Road closure
Measure: Crash risk
•
•
Approach: Friction (or slipperiness)
Approach: Reported crashes
Although 15 measures of winter storm severity were found in the literature, none of the
responding agencies reported using a storm index to normalize the severity of each storm other
than rate of snowfall. Pursuing an operational storm severity index that can be applied to
normalize any other measure over time would be desirable. For example, some performance
measures many be collect continuously through a storm (e.g., friction or traffic density and
speed), while others maybe collected following each storm (e.g., time till bare pavement, crashes
per storm), and others are calculated per season (e.g., all winter crashes, materials used per year,
number of times and duration of road closures); however, winter weather severity indexes are
seasonal, resulting in disparity in the time frames of each and, hence, the usefulness of
performance measurement.
Seventeen agencies reported using customer satisfaction as a measure of performance.
Satisfaction sets the level of performance that the public expects. The performance measures that
were reviewed measure how close winter road maintenance comes to meeting public
expectations. Most agencies use a periodic survey to determine public expectations, and some
track complaints and 511 calls. Best practices for determining customer satisfaction and linking
operational performance to those expectations should be documented, as they are in Chapter 6.
1.6. Conclusions
It is expected that more winter maintenance agencies will adopt performance measurement
practices and that the public will continue to expect clear roads and less harm to the environment
from snow ice control operations. Technologies such as automated vehicle location (AVL),
global positioning systems (GPS), friction meters, road weather information systems (RWIS),
among others, will facilitate obtaining the additional data needed to enhance measuring
performance. The expanded use of these technologies and their increased production and
competition will lead to lower costs. However, both field personnel and management would have
to focus more on outcomes when using these technologies.
7
The objectives selected by each agency can also drive performance measurement by creating
targets toward which activities can be directed. In addition to objectives, performance measures
need to include a short-term result, an improvement strategy, and hold entities accountable.
However, success with performance measurement will require responsive data systems capable
of generating timely data.
Performance measurement offers a promise of improved management and improved outcomes. It
builds on a long history and extensive experience in techniques to strengthen and improve winter
maintenance operations. Developing performance measurement, will lead to more effective
winter maintenance programs.
CHAPTER 2. LITERATURE REVIEW
An international literature review of published and in-progress materials related to snow and ice
control performance measures was conducted as one of the initial tasks under this research
effort. This review involved searches of the Transportation Libraries Catalog, TRIS online,
Transportation Research Board in-progress research, and the Internet.
2.1. Summary of Key Points
The literature review revealed that a significant amount of published materials dealt with
different types of performance measures, both in-use and theoretical but a limited amount of
literature documents agencies’ use of performance measures in day-to-day practice. Various
instances of research and testing of proposed performance measures were described, but often
without implementation or field testing by state or local agencies in the United States. It
appeared that some European countries and Japan are more advanced in developing and
implementing winter maintenance performance measures, possibly because more snow and ice
control operations are contracted to private companies internationally than in the United States.
Also perplexing is the variety of measures that agencies use as winter maintenance performance
measures. Measures such as friction are continuously measured, others are measured per storm,
and still others are measured per season (e.g., number of road closures). Agencies consider each
performance measure collected at varying time intervals and make management decisions based
on each interval and have different expectations of their performance measures.
This review discovered the following:
•
•
•
Three scanning review teams of U.S. officials visited Europe and Japan in 1994,
1998, and 2002, focusing on winter maintenance activities and advanced intelligent
transportation systems (ITS) technologies.
Performance measures can be divided into three general categories: input, output, and
outcome measures
Performance measurement can be collected over several time intervals but are
generally measure continuously, storm-by-storm, and season by season, other
8
•
•
•
•
•
•
•
•
intervals my include week-by-week or month-by-month.
Known input measures include labor hours, equipment hours, various material units,
and expenditures.
Known output measures include cost determined by a unit of accomplishment of
work performed (e.g., lane-miles plowed or sanded), material application rates,
equipping and calibrating trucks, and route characteristics. These measures may also
be based on time and storm event.
Known outcome measures include bare pavement regain time, friction (skid
resistance by coefficient of friction), reduction in crashes, duration and frequency of
closures, advanced warning time to customers, and customer satisfaction (indicated
by customer satisfaction surveys).
A Pavement Snow and Ice Condition (PSIC) chart, as used by some agencies, assists
with uniform pavement condition identification by combining traffic flow
characteristics and visual observation. (Blackburn, et.al 2004)
Various outcome measures can and, are often combined to form an overall Level of
Service (LOS) rating for a roadway.
Contracts with private sector operators are often written such that reimbursement is
based on a combination of input (pay items) and output or outcome measures (with
expectations).
Innovative technologies installed on winter maintenance vehicles that aid in the
collection of data applied to performance measures include AVL, GPS, friction
meters, and various sensors of material, equipment, and temperature.
Winter weather severity indices have been developed to help quantify the relationship
between the severity of winter weather events and roadway condition or safety
factors. However, these indexes are generally lacking because their duration of data
collection is too long for the making of storm-to-storm decisions.
2.2. Background
Development of Performance Measures in the United States
Several U.S. state transportation agencies are utilizing performance measures or standards for
internal assessments or contract monitoring of winter maintenance activities. Performance
measures reflect unique characteristics for each agency, such as the following (TransTech
Management 2003):
•
•
•
•
•
•
•
Agency goals, objectives, and strategies
Organizational and legislative structures and responsibilities
Project development processes
Geography and climate
Fiscal constraints
Rural versus urban focus
Stakeholder concerns
9
The establishment of goals can lead to identifying potential performance measures specific to an
agency’s needs. Goals help relate system performance to reflect what a user perceives the system
should be achieving. Possible goals for winter maintenance include mobility, quality of life,
environmental and resource conservation, safety, operational efficiency, and system condition
and performance (Adams et al. 2003). Adams et al. described a process for identifying and
developing performance measures:
1.
2.
3.
4.
5.
6.
Identify the applications areas for the winter maintenance vehicle data.
Form working group of agency personnel responsible for winter maintenance.
Identify goals and objectives for each application.
Identify the performance measures for the objectives.
Formulate the performance measures using winter maintenance vehicle data.
Identify and develop analytical tools.
Adams et al. (2003) also described principles to guide selecting performance measures:
•
•
•
•
•
•
•
Each measure should be meaningful and appropriate to the needs and concerns of
decision makers.
Each measure reflects specific goals or compliance with guidelines.
The measures reflect current issues, such as environmental concerns.
The measures facilitate comparisons among alternative equipment and operational
strategies for providing the service.
The measures facilitate the prediction of future performance trends for planning and
budgeting.
The measures facilitate the asset valuation and depreciation of equipment.
The measures facilitate comparisons of performance across districts, counties, and
patrol sections.
Scanning Review of International Practices
Some international agencies have been using performance measures in winter maintenance
operations for many years. In an effort for U.S. agencies to improve efficiency and customer
satisfaction, scanning teams were sent abroad to document the process and operations of
international winter maintenance agencies. In the past decade, three teams of U.S. transportation
experts have traveled overseas to study how other countries handle winter maintenance
operations. The first two scans, in 1994 and 1998, primarily focused on maintenance; the third,
concluded in 2002, focused on advanced technologies (Pisano 2004).
The first two tours visited European countries and Japan. The scanning tour discovered that
Japan and many European countries use a “systems concept” that addresses the vehicle, driver,
and the equipment, along with the practices for managing roadway and bridge snow and ice
control. The goals for a systems concept include sustaining or improving levels of winter
maintenance service with the greatest benefit/cost improvements, increasing the safety of winter
driving, and providing an improved level of environmental protection (Smithson 1998).
10
NCHRP Project 20-7, Task 71, utilized two phases in developing a winter maintenance program.
Phase one concentrated on organizing a committee to develop a winter maintenance guide. The
second phase established a Snow and Ice Cooperative Program. In 1997, committee members
identified ten topics as being of high priority that fit into four broad categories: training,
materials applications and specifications, technological advancements, and public
relations/communications. The topics of relevance included the development and validation of
test methods for anti-icing and deicing materials, the investigation of the use and application of
GPS equipment and technology in winter maintenance operations, the measurement of friction
on highway pavements during winter activities, and the investigation and evaluation of
opportunities for using computerized controls and onboard interactive display services in snow
and ice control.
Several European countries are moving toward privatizing winter maintenance operations.
Because of the need to define the expected wintertime LOS, agencies are developing methods to
evaluate the performance operations (Pisano, 2004). In the 1998 scanning tour which visited
France, Switzerland, Norway and Sweden, found that these countries used a bare pavement
policy (Smithson 1998). The performance measures are all measurable and are primarily outputbased. The policy involves the following features:
1. Maximum deterioration of road conditions tolerated before needed action
2. Minimum LOS conditions before action is ended
3. Time frames for achieving the LOS based on weather conditions
2.3. Types of Performance Measures
A combination of various input, output, and outcome measures may be combined to determine a
LOS of the winter maintenance operations. The three categories are discussed further in the
following sections.
Input Measures
Input measurements are used to quantify the resources spent on snow and ice control or winter
maintenance operations, typically applied to equipment, material, and labor used for winter
operations. Quantifying this value is done in terms such as number of trucks, labor-hours, and
volume or tonnage of material. During the operation, the number of equipment amounts and
labor-hours may be documented. In contracts, the pay items are directly related to time and
usage, based on the quantified measures desired (Bourdon 2001).
Output Measures
Output measures quantify physical outputs from the resources that are used in units of work of
winter operations. Output specifications primarily deal with defining methods of performing the
work and the associated accomplishments. In contract specifications, pay items may be based on
route characteristics, storm events, truck operations, truck characteristics, and time (Bourdon,
2001). Cost is one of the most common measures used to establish a performance-based system
11
established by a unit of accomplishment at the crew level (AASHTO, 1999). Examples of route
characteristics include a specified order of plowing roads, development and performance within
a plowing network, or number of rounds needed during an event. When compared to time, lanemiles per unit of time sanded or plowed are measured. Truck operations include plowing speed
and material application rates. Payments for winterizing, calibrating, and equipping trucks
specifically for winter operations are also examples of output-based pay measures.
Outcome Measures
Outcome measures reflect the end result of winter maintenance during and after a storm event,
usually as perceived by the motorist. The user of a road typically has expectations on how a road
should handle, thus relating the performance of the maintenance to what the motorist feels, sees,
and expects in terms of recovery time (Bourdon 2001). The user also wants access and mobility
for unrestricted and safe travel.
Outcome measures reflect an agency’s success in meeting goals and objectives, typically from
the customer’s perspective. Common types of outcome measures are (Bourdon, 2001, Blackburn
et al. 2004) include the following:
•
•
•
•
•
•
•
Bare pavement regain time
Friction (skid resistance)
Reduction of crashes
Duration and frequency of closures
Advanced warning time to customers,
Customer satisfaction surveys
Visual characteristics
The most popular choices are bare pavement regain time, friction testing, and customer
satisfaction surveys, which take into account a driver’s visual and physical perception of the
roadway surface (Blackburn et al. 2004). These outcome measures are discussed in the following
subsections.
Visual and Physical Perception of Roadway Surface Conditions
Visual characteristics of road conditions are of greatest concern to the motorists. Road conditions
can be assessed by different measurements, visual and physical. Unevenness, rutting, and
slippery conditions are concerns of drivers during and after a storm event. Visual characteristics
are easily identifiable without physical testing. Different visual roadway characteristics include
(Blackburn et al. 2004) the following:
•
•
•
•
Centerline bare
Wheel path bare
Loose snow covered (percent area and depth)
Packed snow covered (percent area and depth)
12
•
•
•
•
•
•
•
•
Bare (percent area)
Thin ice covered (percent area)
Thick ice covered (percent area)
Dry
Damp
Slush (percent area and depth)
Frost
Wet
A Pavement Snow and Ice Condition (PSIC) table (Blackburn et al. 2004) is developed by using
visual characterization of roadway surfaces together with traffic flow and other visual
information to identify a level of service of the road. The PSIC table correlates what a driver
would perceive the condition of a road to what the driver sees and how the driver feels while
driving on the road. The PSIC allows for identifying a distinct condition of the roadway with
relevant, useful information to the agency and motorist. The PSIC helps an agency determine
what method of maintenance is desired for effective winter maintenance and the instantaneous
visual status of measured outcomes. Table 2 shows a sample PSIC table.
13
Table 1. Sample PSIC table
Conditions
Description
Condition 1
All snow and ice are prevented from bonding and accumulating on the road
surface. Bare/wet pavement surface is maintained at all times. Traffic does not
experience weather-related delays other than those associated with wet
pavement surfaces, reduced visibility, incidents, and "normal" congestion.
Condition 2
Bare/wet pavement surface is the general condition. There are occasional areas
having snow or ice accumulations resulting from drifting, sheltering, cold spots,
frozen melt-water, etc. Prudent speed reduction and general minor delays are
associated with traversing those areas.
Condition 3
Accumulations of loose snow or slush ranging up to (2 in.) are found on the
pavement surface. Packed and bonded snow and ice are not present. There are
some moderate delays due to a general speed reduction. However, the roads are
passable at all times.
Condition 4
The pavement surface has continuous stretches of packed snow with or without
loose snow on top of the packed snow or ice. Wheel tracks may range from
bare/wet to having up to (1.5 in.) of slush or unpacked snow. On multilane
highways, only one lane will exhibit these pavement surface conditions. The
use of snow tires is recommended to the public. There is a reduction in traveling
speed and moderate delays due to reduced capacity. However, the roads are
passable.
Condition 5
The pavement surface is completely covered with packed snow and ice that has
been treated with abrasives or abrasive/chemical mixtures. There may be loose
snow of up to (2 in.) on top of the packed surface. The use of snow tires is
required. Chains and/or four-wheel drive may also be required. Traveling speed
is significantly reduced and there are general moderate delays with some
incidental sever delays.
Condition 6
The pavement surface is covered with a significant buildup of packed snow and
ice that has not been treated with abrasives or abrasives/chemical mixtures.
There may be (2 in.) of loose or wind-transported snow on top of the packed
surface due to high snowfall rate and/or wind. There may be deep ruts in the
packed snow and ice that may have been treated with chemicals, abrasives, or
abrasives/chemical mixtures. The use of snow tires is the minimum
requirement. Chains and snow tire equipped four-wheel drive are required in
these circumstances. Travelers experience severe delays and low travel speeds
due to reduced visibility, unplowed loose, or wind-compacted snow, or ruts in
the packed snow and ice.
Condition 7
The road is temporarily closed. This may be the result of severe weather (low
visibility, etc.) or road conditions (e.g., drifting, excessive unplowed snow,
avalanche potential or actuality, glare ice, accidents, vehicles stuck on the
road).
14
A 2001 survey of business owners conducted in Sapporo, Japan, was used to evaluate traffic
issues during winter travel. Three highly ranked problems, from a business owner’s perspective,
were risk of increased traffic accidents, decline in visits to clients for meetings/sale, and increase
in time to deliver merchandise. Three relationships were developed based on the responses
(Yamamoto et al. 2004):
•
•
•
“Increased risk of winter accidents” is largely influenced by the skid resistance
coefficient of the winter road surface.
“Increase in time to deliver merchandise” is strongly influenced by reduced traffic
capacity from narrowed road width.
“Decline in visits to clients for meetings or sales” is strongly influenced by both the
increased risk of accident and the increased time required traveling to clients.
Addressing this issues means improving the coefficient of friction of winter road
surface and maintaining the road width.
The five most important indicators of winter road maintenance are road surface conditions (i.e.,
unevenness, rutting, and slipperiness), road width, sight distance at intersections, pedestrian
safety, and provision of traffic congestion information. The top ranked indicators were road
width, followed by road surface conditions (Yamamoto et al. 2004).
In the study by Yamamoto et al., outcome indicators were designated as maintained road width
and coefficient of friction. For these indicators, quantitative goals, targets of fulfillment rates,
and the comparison of the target achievement rates to the actual achievement rates were set as
measurements for the outcome.
Bare Pavement Regain Time
Bareness of pavement is a performance measure for winter maintenance used during and after a
storm event that is understood by users. Two definitions of bare pavement are acceptable by
drivers (Bourdon 2001):
•
•
Bare pavement. Driving lanes are bare with centerline and edge lines showing.
Bare lanes. Driving lanes bare with centerline and edge lines covered.
Mn/DOT conducted market research in 1994, 1996, and 2000 to determine the pavement
conditions at which the public was satisfied. The research indicated that the public felt that the
time to obtain an adequate LOS is after a storm was also important. A high public satisfaction
was reported at 90% bare lane (Keranen 2002).
The concept of “bare pavement regain time” was originally set by Mn/DOT at 95% of the
roadway clear of ice and snow, but was changed to 90% clear of ice and snow in the following
combinations (Keranen 2002):
•
•
Ten 50-foot spots per mile
Two 250-foot spots per mile
15
•
•
•
Two mile-long spots per 20 miles
Two half-mile spots per 10 miles
Eight quarter-mile spots per 20 miles
In 2001, a study was performed to determine the roadway conditions that motorists would find
acceptable for driving to and from work or to other daily appointments (Niemi 2001). The
recommendation from this research was that the greatest impact on improving roadway
conditions is to first clear two lanes on all road classes before upgrading roadways to fully bare.
In the rural Minnesota area, the study recommends clearing four-lane roadways to make the
travel way clear, with the centerline covered, with or without the edge lines showing, before
upgrading the four-lane roads to fully bare. Within the Minneapolis/St. Paul metro area,
interstate highways and four-lane roads should be cleared to two lanes, with centerline and edge
lines covered, by 6 a.m., and two-lane roads should be cleared to this condition by 7–9 a.m. for
greatest accessibility. In afternoon driving, an improvement over one intermittent wheelpath is
desired on two-lane roads, and the interstate highways should be fully bare if possible (Niemi
2001).
Roadway Friction
The Yamamoto et al. study (2004) also found that securing adequate friction to be another
important aspect of winter maintenance. The coefficient of friction may be increased with deicing, anti-icing, and sanding the roadway. Friction testing is a way to measure the effectiveness
of this performance. Norway proposed the use of 0.25 as the coefficient of friction threshold
when spreading sand on snow packed roads (Al-Qadi et al. 2002).
In the United States, friction meters have been primarily installed on vehicles for testing and
research, while some European countries use them in operational applications. Three methods of
friction measurements include a model-based approach using climate, traffic, and roadway
conditions to predict friction; direct friction measurements by an extra wheel installed on
vehicles; or by using traction control systems.
Al-Qadi et al. (2002) suggested four scenarios concerning the methods of friction testing:
1. Friction measurements by a winter maintenance patrol vehicle
2. Friction measurements by winter maintenance snowplow/spreader vehicles
3. Recorded, archived friction measurements by winter maintenance patrol or
snowplow/spreader vehicles
4. Recorded, archived, and real-time transmitted friction measurements by winter
maintenance patrol or snowplow/spreader vehicles
In the 1998 Nixon study, the author suggests three operational uses of friction measuring
devices. First, these devices may be used as a measure of quality, which is beneficial to agencies
that need a tool to measure the performance of winter maintenance contractors. Second, friction
devices may be used as a source of road user information to inform motorists of hazardous
locations of roadways with low friction. Third, friction devices could be used as a means of
16
controlling chemical application by determining the amount of chemical de-icing to be placed on
the roadway.
2.4. Performance-Based Levels of Service
Many performance measures are being tried and related to an overall LOS. According to
(Blackburn et al. 2004), the most popular measures are the following:
•
•
•
•
Pavement conditions (visual)
Performance indices relating amount of time pavement is covered in snow/ice to
storm total (visual)
Report cards (customer satisfaction surveys)
Friction measurements and slipperiness ratings
When relating these measures to LOS in winter maintenance, the primary considerations are
cycle time, available material treatments, weather and site conditions, and traffic considerations
(Blackburn et al. 2004). Achievable LOS ratings are dependent on the average daily traffic
(ADT) volumes of the facility and the capabilities of the agency. Roadway function levels often
determine the type of treatment that will be applied to a facility in a specific type of weather
event.
Agencies winter maintenance capabilities differ in terms of equipment, labor, and materials used
for application. Two timeframes are to be considered when measuring performance-based LOS:
within-winter weather event and after-end-of-winter weather event.
Two components compose a within-winter weather event, the amount of loose snow/ice/slush
that is allowed to accumulate between plowing cycles and the condition of the ice/pavement
interface in terms of bond and packed snow/ice. Often, plowing resource requirements is
governed by the amount of loose snow allowed to accumulate on the road surface between
plowing cycles (e.g., equipment resources). The plowing production rate is combined with the
design snowfall rate to yield a cycle time required to meet and “accumulation” goal. The
condition of snow/ice pavement interface in terms of bond or packed snow/ice is a function of
pavement temperature, type of treatment, treatment application rate, and cycle time.
The time to achieve particular pavement surface conditions in terms of ice or snow coverage, or
PSIC level, is expressed as the after-end-of-winter weather event LOS. Ratings are usually color
coded or translated into letter designations A, B, C, D, and F (Blackburn et al. 2004).
2.5. Performance-Based Pay Items in Contracts
Contracts may allow the reimbursement to the contractor is based on the consumption of inputs,
production of outputs, or the delivery of outcomes. Many contracts are a blend of inputs (pay
items) with output or outcome levels (expectations). Input-based pay items are directly related to
time and usage. The unit cost rate may be based on labor hours, equipment hours, or material
17
used. Output based pay items are related to work accomplishments in units of work performed.
Outcome based pay is usually lump sum payments for a season of winter maintenance for a
specified location (Bourdon 2001).
2.6. Innovative Use of Technology for Performance Measurement
It is becoming more common for winter maintenance vehicles to be equipped with technology
for measuring performance. In Wisconsin, for example, AVL, GPS, material sensors, equipment
sensors, and temperature sensors are some of the more advanced technologies that collect data to
be used in improving the winter maintenance process and operational methods (Adams et al.
2003). Real-time data of material application rates, location, equipment status, and roadway
characteristics allow for instant operational decision making along with post-operational analysis
and summaries.
In 1999, the Wisconsin Department of Transportation installed AVL, material usage sensors,
front and wing plow status sensors, under-body scraper sensors, and air and pavement
temperature sensors to record data as often as every two seconds. Other state DOTs, including
Iowa, Michigan, and Minnesota, have worked with Iowa State University and private vendors to
equip vehicles with AVL, air and temperature sensors, friction meters, anti-icing and pre-wetting
equipment, equipment status sensors, reverse obstacle sensors, and in-vehicle heads-up displays
of sensor data (Adams et al. 2003).
The scanning review teams documented many international implementations of sensors. Japan
uses ground-view sensors to monitor eight road surface conditions with the eventual goal of
being able to adjust chemical applications automatically. Italy uses GPS and AVL technologies
to assist with programming variations in chemical application and tracking and billing of
materials the spreader has placed (Pisano 2004).
Research is being conducted on the use of automatic traffic recorders to record vehicle speeds to
develop speed recovery time as a performance measure (Lee and Ran 2004). This research will
combine the average vehicle speed reduction during a winter storm event with storm report data
to determine the time needed to regain the normal vehicle speeds.
2.7. Winter Weather Severity Indexes
A roadway weather severity index is used in the road weather community to quantify the
relationship between winter weather severity and roadway conditions or safety. Most weather
severity indexes provide only measures of relative severity of an entire seasons and this only
allows a seasonal comparison of the relative severity of winter. Several winter severity indices
developed for general or for specific purposes are described as the following:
Hulme and Modified Hulme Index
Hulme made one of the first attempts to develop a winter index to numerically classify winter
severity. A winter was defined as the time between the first of December and the end of March
18
(Hulme 1982). This index uses three parameters in the winter index computation: the mean daily
maximum temperature, the number of days with snow lying at 9:00 a.m., and the number of
night ground frosts (Hulme 1982). A constant was also added into the equation to ensure that the
weather index averages to zero. The original and modified indices are expressed by the
following equations:
Winter Index: WI=10T-18.5S-F+200
Modified Winter Index: WI=10T-(18.5S)1/3-F+C
A low index value indicates a severe winter, while a high number indicates a mild winter that
can only be used to summarize seasonal weather.
Pennsylvania Department of Transportation Index
Rissel and Scott (1985) developed a winter severity index for the Pennsylvania DOT based on
total meteorological data to relate the severity of winters to labor costs to help establish optimum
winter staffing patterns and determine staffing cost-effectiveness. The index is presented in the
following equation:
SI = S + 2M + H + T – (C/2) + R
Where:
•
•
•
•
•
•
(S) is the total inches of snowfall in a period.
(M) is the number of days with snowfall of 1 to 6 in.
(H) is number of days with snowfall greater than 6 in.
(T) is the number of days with a maximum temperature above 32 degrees F and a
minimum temperature below 32 degrees F.
(C) is the number of days with temperatures below 32 degrees F.
(R) is the total hours in the period when snow or ice occurs.
Strategic Highway Research Program Index
Boselly et al. (1993) developed as part of the Strategic Highway Research Program, SHRP
Project-H-350, a winter index that quantitates expression of winter severity. The index is defined
by the following four equations:
⎛S
⎞
⎛ N ⎞
WI = −25.58 TI − 35.68 ln⎜ + 1⎟ − 99.5 ⎜
⎟ + 50
⎝ 10 ⎠
⎝ R + 10 ⎠
Where:
19
•
•
•
•
TI is the average daily temperature index
S is the mean daily snowfall
N is the mean daily values of number of days with air frosts
R is the temperature range
⎛S
⎞
⎛ N ⎞
WI = a TI − b ln⎜ + 1⎟ − c ⎜
⎟+d
⎝ 10 ⎠
⎝ R + 10 ⎠
This is the index expressed in a general form where a, b, c, and d are coefficients that reflect are
particular weights and critical values of the parameters in each term, for typical weather
conditions encountered at a given location. In this study, the variables were weighted to account
for the critically significant level of each parameter to winter maintenance cost.
The Kansas DOT and Mn/DOT have adopted this index, and Ontario has modified the equation
to include freezing rain by adding the number of freezing days (McCullouch et al. 2004).
University of Waterloo Index for Ontario Highways on Salt Use
Audrey et al. (2001) conducted a study to assess the suitability of indices developed by Boselly,
et al.; Hulme; and Salt Day, which explains the temporal and spatial variability of salt use on
highways in Ontario, Canada. The indices were modified to reflect Ontario’s climate. The frost
term in the SHRP model was replaced with a freezing rain indicator to better represent the
variability in monthly salt use. The SHRP model was found best suited for Ontario because it
places the most emphasis on snowfall. The adapted model is presented by the following:
⎡S
⎤
WI = −25.39 TI − 23.27 ln ⎢ + 1⎥ − 99.5 ( frz ) + 50
⎣10 ⎦
Where:
1. TI is the average daily temperature index
2. S is the mean daily snowfall
3. Frz is freezing rain indicator
Knudson Developed Index in Denmark
Knudsen (1994) developed a winter index in Denmark for every day and county, that considers
road temperature, the number of road freezes in a day, snowfall, and presence of snow drift. The
index is presented as follows:
WI =
Apr 15
∑WI
Day
Oct 15
WI Day = x freeze (1 + x frost + x refreeze + x snow + x drift )
20
This index did not reveal a relationship between winter maintenance expenditures and index
values, but the relationship between the index and salt consumption was R2=0.96 on a seasonal
basis.
Indiana Winter Severity Index
McCullouch et al. (2004) developed a winter severity index for the Indiana DOT based on
surveys of field crew and employees to identify the weather factors that had the most influence
on the snow and ice removal effort. The survey factors were the number of days with
temperature and dew point below freezing, the number of days with freezing rain, the number of
“snow event days”, and the number of days with drifting snow. To improve correlation within
the model, three additional factors were included in the equation. They were snow depth, snow
duration, and average temperature.
To account for regional climate differences within the state, separate equations were developed
for each zone. The following statewide equation was developed:
WI=0.71839*Frost+16.87634*FreezingRain+12.90112*Drifting-0.32281*Snow+
25.72981*Snow Depth+3.23541*Hour-2.80668*Average Temperature
Iowa State University/Iowa DOT Index
Carmichael et al. (2004) developed a winter weather index for estimating winter roadway
maintenance costs in the Midwest. To relate to winter maintenance costs and weather parameters
the index utilized both regression analysis and neural networks for correlation calculations
concerning the following:
•
•
•
•
Precipitation
Temperature
Date
Wind
To factor in dependent cost variables, operations data were also used. The index is used to judge
how well the maintenance personnel performed statewide each winter season by estimating what
costs should have been incurred and the amount of hours that should have been used in winter
maintenance.
Wisconsin Winter Severity Index
The Winter Severity Index used in Wisconsin to evaluate the counties’ performance on snow and
ice removal, is expressed by the following equation (McCullouch et al. 2004):
21
WI = 10 *
SE
FR
AMT
DUR
INC
+ 5.9 *
+ 8.5 *
+ 9.4 *
+ 9.2 *
63
21
314
1125
50
Where:
1.
2.
3.
4.
5.
SE is snow events
FR is freezing rain events
AMT is snow amount
DUR is storm duration
INC is incidents (including drifting, cleanup, and frost runs)
Nixon and Qiu Index
Nixon and Qiu (2004) developed a storm severity index based upon that developed by Boselly,
et.al., in the SHRP Project H-350, to determine to what extent an individual storm poses
difficulty to maintenance activities. This index is unique in that provides a measure of severity
for any given storm based on meteorological data. The utility of this index is but one step in the
process of creating a quality controlled winter maintenance program. This index provides
agencies with storm by information thus; it can be used to measure the performance of a given
agency in handling a given storm and as such represents an important part of a quality control
process for winter maintenance. The index uses a matrix of possible storms to classify events.
The equation is expressed as follows:
⎡1
⎤
SSI = ⎢ * [(ST * Ti * Wi ) + Bi + Tp + Wp − a ]⎥
⎣b
⎦
0.5
Where:
1.
2.
3.
4.
5.
6.
7.
ST describes storm type
Ti is the in-storm road surface temperature )
Wi is the in-storm wind condition
Bi is the early storm behavior
Tp is the post storm temperature
Wp is the post storm wind condition and
A an B are parameters to normalize the storm severity index from 0 to 1
Transportation Association of Canada Index
According to its website, the Transportation Association of Canada (2005) is currently
developing a winter severity index that will allow the forecasting of the relative harshness of a
given winter compared to a base year for each province and territory. The index will be based on
22
an assessment of other existing indicators used throughout the world and their applicability for
use in Canada, while tailored to each jurisdiction.
Washington State Department of Transportation Index
The Washington State Department of Transportation (WSDOT) developed a frost index, which
was found to directly relate performance measurement in winter activities. The frost index is a
severity index that does not include the snowfall factor and was planned for use to justify an
overrun in the snow and ice budget in case of supplemental funding (Perry and Symons 1991).
Strong and Shvetsov Index
Strong and Shvetsov (2005) used data from California, Montana, and Oregon to quantify
relationships between winter weather and safety. Linear models were developed for different
topographic zones and statewide to predict the cubic root of the crash rate as a function of
weather parameters. The models incorporated weather data from National Weather Service
stations, crash data within five miles of the weather stations, and annual ADT volumes with
monthly adjustment factors.
The Salt Day Indicator
A “salt day indicator” was developed by the Illinois State Water Survey and is being used by the
Illinois DOT. The index is a count of a number of days within a month that meet specific criteria
for snow removal budget allocations through short-term forecasts (Cohen 1981). The equation is
expressed as follows:
WI=Dsnow + Dcold
Where:
Dsnow= Daily snowfall accumulation is greater than or equal to 0.5 in.
Dcold= Number of days where mean daily temperature is between 15° and 30°F
Road Sense Index
The Road Sense Index sponsored by the Insurance Corporation of British Columbia to estimate
winter driving risk for the greater Vancouver metropolitan area, by correlating weather
parameters with safety. It was intended to alert motorists about hazardous conditions so drivers
can adjust their driving behavior (Chen et al. 1994).
23
Decker Index
This index was to develop a measure of winter maintenance efficiency, accounting by
considering labor, equipment and material costs as influenced by storm severity and duration to
achieve a specific number of lane-kilometers of given LOS. The index was based on the SHRP
equation, and is expressed by the following (Decker et al. 2001):
⎛S
⎞
⎛ N ⎞
WI = −25.59 TI + −11.50 ln⎜ + 1⎟ − 99.50 ⎜
⎟ + 50
⎝ 10 ⎠
⎝ R + 10 ⎠
CHAPTER 3. WINTER MAINTENANCE MEASURES USED BY HIGHWAY
AGENCIES
This chapter identifies and discusses the winter maintenance performance measures used by
several states, provinces, cities, and counties. There is a broad range of uses of performance
measurement, ranging from not measuring performance of snow and ice control at all to those
exercising sophisticated measures of performance of operations. The landscape of performance
measurement is wide ranging. While many agencies stated the need for performance
measurement, only a few of these have established a formal performance measurement process
for their operations. In general, the agencies have focused their efforts on achieving the desired
results of effective snow and ice control to meet the demands of the traveling public.
3.1. Survey Results
In this project, a survey was sent to 162 winter maintenance operations personnel including some
in other countries. The targeted survey respondents were from local, state, and federal agencies.
The respondents were chosen to provide feedback unique to their areas of expertise.
Of the 162 surveys distributed, 43 surveys were completed included responses from state DOTs,
four Canadian provinces, one response from Sweden, one from Japan, one from the City of
Edmonton AB, and 17 from cities and counties in the U.S. Table 2 lists the agencies that
responded to the survey. The responses provided insights into the use of performance measures
in winter maintenance operations, particularly in the northern hemisphere regions. The map in
Figure 3 indicates the jurisdictions that responded to the survey. The respondents were primarily
from the United States and Canada.
The specific responses from the 43 agencies are found in Appendix B, and the findings of the
survey are presented in this chapter.
24
Table 2. Locations responding to the survey
U.S. state agencies
responding
Alaska DOT
U.S. cities/counties responding
Provinces/countries responding
Ada County Highway District, ID
Manitoba, Canada
Washington DOT
Detroit, MI, Public Works
Department
Ontario, Canada
Minneapolis, MN, Public Works
Department
Saskatchewan, Canada
Des Moines, IA, Public Works
Department
Canadian cities responding
California (CalTrans)
Arizona DOT
New Mexico DOT
Nevada DOT
Colorado DOT
Kansas DOT
Nebraska DOR
Iowa DOT
Jackson County, MO, Public
Works Department
Missouri DOT
Wisconsin DOT
King County, WA
Illinois DOT
El Paso County, CO, Department
of Transportation
Ohio DOT
New York DOT
New Hampshire DOT
Maryland SHA
Minnesota DOT
City of Edmonton, AB
Erie County, NY, Public Works
Department
Indianapolis, IN, Public Works
Department
Indiana DOT
Alberta, Canada
Other countries responding
Swedish National Road
Administration
Japan
Cedar Rapids, IA, Street
Department
Mc Henry County, IL, Division of
Transportation
Seattle, WA, Department of
Transportation
Douglas County, NE
West Des Moines, IA
Washington County, MN
Cook County, IL, Highway
Department
Cuyahoga County, OH,
Engineer’s Office
Performance Measures
The respondents identified the methods used for conducting snow and ice removal operations.
As Figure 2 indicates, the majority of respondents, (65%), are using their own staff for snow and
ice control. The remaining states or provincial agencies responded that they contract with others
25
to perform the operations, including private contractors or local agencies to plow selected routes
performing all snow and ice control operations. Outsourcing of this work is usually done to
reduce expenditures.
Number of Responses
Personnel Used for Snow and Ice Control
38
36
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Internal Staff
Private Contractors
Contracts with
other agencies
Type of Personnel
Figure 2. Personnel used for snow and ice control
The respondents who stated that they contracted out snow and ice control were asked if an
evaluation of the contractors’ performance was in place. In most instances, the contracting
agency evaluates the contractors’ performance to ensure that the contractor is meeting its
obligations. However, four respondents indicated that they do not evaluate contractors’
performance. Evaluating contractor’s performance helps in to meeting the traveling public’s
expectations for clearing the roads in a reasonable amount of time. One example is evaluating
the contractor’s performance is the Virginia DOT (see Figure 3). VDOT has the contractor
specify, as much as possible, the measurable outcome to be achieved. VDOT also requires
contractors to prepare their own Snow and Ice Plan as to what resources will be used and how
they will be used to achieve that outcome. Finally, VDOT require that the contractor’s Snow and
Ice Plan be approved by the owner-agency prior to the contract award.
26
Number of Responses
Do you evaluate others performance?
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Yes
No
Not Applicable
Responses
Figure 3. Evaluation of performance
The responding agencies described the relationship of contractor performance to the payment for
work done. In eight instances, the contractor’s performance is linked to payment, as indicated in
Figure 4. However, in 11 cases, no formal method to tie the contractor’s performance to payment
was reported.
27
Is performance linked to payment?
20
Number of Responses
18
16
14
12
All
State
Local
10
8
6
4
2
0
Yes
No
Not Applicable
Responses
Figure 4. Performance and payment
28
Thirty-three agencies, including 16 state and provincial agencies and 17 local agencies reported
that they measure performance in snow and ice control operations. Figure 5 indicates the
performance measures used by the respondents.
Measures Identified
26
24
Number Responses
22
20
18
16
All
State
Local
14
12
10
8
6
4
2
Ti
m
Ti
m
e
Ti to
Ti e to m e b ar
m r
e
e e t to w p a
to u r
v
e
n
p r t t p em
ov o
av en
id ne em t
e ar
e
1
wh n o n t
ee r m a
Le
l
l
v e F t r ac
r
l o ic k
C Cu
t
r
s
T f S io
T r as h t om r a e r n
a f e s e ve vi c
f ic p r s l s e
v e a p
T i o l u r v e t i sf ee d
m m h ac
e e ic ti
fo du le on
rt r m
r a i ng i l e
La
ffi s s
ne
c to
-n r
m
ile Fu o r m
s e m
Pe ( km l us al
Am
rs ) ag
o p
e
o T
M un o n s O v n n e l o w
ile t o
e
e
o r lh d
s
(k f e q f m tim ou
m u at e rs
) ip er ho
C t r a m ia u
% o s v e en t l s u r s
sa t o led d e se
lt/ f o - p l d
s p p pl o y
r e er a ow e d
a d ti d
e r on o w
s s n
c a ln
lib / m
ra i
t
O ed
th
er
0
Performance Measures Used
Figure 5. Performance measures used by various agencies
A number of the measures identified are traditional measures that are being tracked, primarily for
budget purposes. However, four agencies are experimenting with other measures such as friction
measuring devices, road closures, snow depth, and number of times tire chains are required. The
following “traditional” measures often cited by local agencies and also by state agencies:
•
•
•
•
•
Fuel usage
Lane-miles plowed
Personnel hours
Overtime hours
Amount of equipment deployed
Other popular measures cited by all levels of agencies are:
29
•
•
•
•
Time to bare pavement
Time to return to near-normal conditions
LOS (mobility)
Customer satisfaction
Of the agencies using friction as measure, Sweden has the most advanced system of
incorporating friction measurement into a performance measurement. The friction coefficient is
determined in accordance with the Swedish National Road Administration (SNRA) methods and
specifications, and “trigger” values are used to determine those areas where the friction is less
than adequate and additional treatment is required. The Ohio DOT is also experimenting with
using friction as a measurement, but at this point using friction is still in the testing phase.
The survey sought information on how the agency decides which items to measure in snow and
ice control operations. Statewide guidelines, committee input, and budget decision processes
were the most common responses. State and provincial agencies, frequently indicated that
statewide guidelines, plans, and policies shape the performance measurements. These plans may
be made by committee with local input, but the states seem to strive toward a statewide standard
of performance measurement. Specific responses from the states included the following:
•
•
•
•
•
Statewide winter operations teams recommend measures, with management approval.
Measured items chosen using existing guidelines and evolving technology.
Data are used that were already being captured (road condition information).
Measured as stated in the agency’s policy and procedure manual.
Performance measures on bare pavement, material usage, and cost of operations are
detailed in the business plan, which is developed by senior managers. Performance
measures for hired equipment are determined by the Statewide Maintenance Quality
Council, which is composed of district and statewide maintenance managers.
Local agencies use of performance measures is budget driven. Local agencies seem to use
traditional, i.e., tried and true, measurements that are required for maintenance management
systems. Local agencies select performance measures with consideration to the following:
•
•
•
•
•
•
Resources and safety
Customer indicators and fiscal barriers
Budget planned versus actual cost for snow and ice; the standard cost per mile versus
actual cost per mile
Materials, personnel, and amount of equipment used
Maintenance management systems outcomes
Decision by department commissioner
Several agencies indicated that they measured inputs and outputs, but have not established any
formal performance measurement process. The input and output measures were tracked for
budgetary reasons; some measures were simply tracked because they “have been historically
tracked” or were to be input into the management system. Many of the agencies are attempting to
30
determine the important items that should be measured and how to determine their effectiveness,
and also how to meet the increasing expectations of the public for bare pavement.
Responding agencies also identified safety and mobility of the traveling public as the most
critical to snow and ice control operations. Twenty-five of the responding agencies indicated
pavement condition or public safety as most critical to their snow and ice control operations. The
amount of time to return the pavement to “normal” driving conditions and to minimize traffic
delays were the focus of the responding agencies. Snow and ice control operations relate to
customer satisfaction, which was mentioned by ten agencies; therefore, agencies at all levels are
striving to strike a balance between the public’s expectation of clear roads and budget
constraints.
Most agencies stated that targets are set annually, usually after the snow and ice season, and the
review of how operations were performed. Some agencies established targets more frequently.
For example, Iowa DOT stated that it sets targets or objectives quarterly; El Paso County,
Colorado, establishes its objectives semiannually; the Ohio DOT sets its objectives as an
ongoing process, the Missouri DOT sets its performance objectives on an “as deemed necessary”
basis. Sweden, which hires contractors for its snow and ice control operations, set performance
objectives any time changes are made in the contract or in the operating rules. Typically, a
contract is set for five years. Figure 6 describes the frequency target setting reported by the
responding agencies.
Number of Responses
How frequently do you set targets?
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Quarterly
Annually
Every 2 years
Other
Frequency
Figure 6. The frequency with which targets are set
31
Providing mobility for the traveling public was cited 25 times by the respondents and ensuring
safety for the traveling public was cited 18 times by the respondents as one of the most important
objectives for snow and ice control operations. Other objectives cited by the respondents
included efficient use of resources, meeting customer satisfaction, and protecting the
environment.
Measures and Performance Level
While many input measures are tracked, usually for budgetary purposes, specific performance
levels are often not established. However, some agencies have established performance measures
and performance levels to ensure proper measurement. Table 3 indicates a sample of measures
and performance levels indicated by the respondents.
Table 3. Performance measures and performance levels
Measure
Time to wet or dry condition; safe travel
way for category I routes
Performance Level
As soon as possible after end of storm
bare/wet wheel paths
Time to clean up in urban areas after storm
event stops
18 hours
Monitor police and public
observational/calls
Bare Pavement
Minimal complaints/calls
Reaching wet or dry pavement within 8 hrs
of the ending of frozen precipitation
Salt
Annual usage
On high volume roadways, return roads to
reasonable, near normal conditions within
24 hours
95%
Follow Maintenance Best Practices (circuit
time) contract equipment complement
Meets theoretical circuit time
Safety
Crash rates
Time to bare pavement
Friction
Time to wet or dry condition
depends on ADT
Friction value
As soon as possible after end of storm
Costs
Snow depth
Budget levels
Centimeters
32
The respondents indicated that roadways with higher ADT receive priority for treatment and
specific measures and performance levels are established for the higher volume roads. The time
frame to achieve the target for the lower traffic volume roads is longer. As Figure 7 illustrates,
most of the data are obtained through observation, either through field observations or from
maintenance workers or law enforcement personnel. Some local agencies use closed circuit
television cameras from the freeway management systems. Post-processing of information is
conducted by some states, in that they obtain data from accounting records to measure how the
agency performed.
Performance measurement data sources
Number of Responses
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
Ac
Vi
co
su
a
t
un
n
lI
in
sp
g
e
All
State
Local
r
o
ec
io
ct
su
Vi
al
I
n
rd
s
by
ns
m
c
pe
ai
tio
nt
n
e
n
na
by
la
ce
w
f
en
or
c
em
en
Re
p
t
ts
or
Ca
fr o
ll s
m
fr o
fie
m
ld
pu
ic
bl
,e
.
5
g.
11
CC
T
F
V-
M
S
Pe
AT
rio
c
di
cu
Rs
s
m
to
er
s
v
ur
ey
s
O
th
er
Data Sources
Figure 7. Data sources for performance measurement
As for other measurements, Ohio DOT and the Swedish National Road Administration (SNRA)
use friction measurements. The SNRA contracts the snow and ice control, and then obtains the
contractors’ records including automated vehicle location (AVL). The contractors must meet
ISO 9000 quality standards. Ontario’s contractors also have AVL in some locations. Through
tracking these vehicles, the contracting agency can better ensure the vehicles are treating
possible problem areas.
The respondents were then asked whether they conducted surveys of the general public about the
agency’s performance of snow and ice control. This is important because several agencies
33
indicated that customer satisfaction was viewed as a critical measure to their performance. The
respondents were evenly divided as to whether they conducted surveys of the general public (see
Figure 8).
Number of Responses
Public Satisfaction Surveys
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Yes
No
N/A
Responses
Figure 8. Agencies using public satisfaction surveys
Agencies that have conducted surveys reported that the public viewed their performance
favorably, although several respondents indicated that the public’s expectations of service are
higher than anticipated. For example, the Colorado DOT indicated that the public rated its
performance between B and B-. Sweden stated that its survey indicated that the public rated its
performance on higher traffic volume roads as good, but poor on lower traffic volume roads.
(This ranking is consistent with the priority placed on higher traffic volume roads.)
About four of the respondents use the same performance measures for roads with different
volumes and surfaces (see Figure 9). About one-third reported that higher traffic volume
roadways receive priority and are treated first, before those roads with lower traffic volumes
such that the higher volume roads are reaching near-normal conditions faster than the lower
traffic volume roads. (A listing of agencies that conducted public satisfaction surveys is included
in Appendix C.)
34
Number of Responses
Performance Measures and Road Characteristics
26
24
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Yes
No
N/A
Responses
Figure 9. Performance measures and road characteristics
As Figure 10 shows, a majority (55%) of the respondents indicated that performance objectives
do not vary with road or storm characteristics, although state agencies were nearly equally
divided. For the most part, local agencies indicated that their objectives did not vary with storm
or road characteristics. The variations of performance objectives were very similar to the
question earlier in that the respondents indicated that the higher traffic volume roads had higher
priority for treatment than lower volume roads. The roads with the higher traffic volumes are to
achieve bare pavement status more quickly than lower traffic volume roads.
35
Do performance targets/objectives used vary with road or
storm characteristics?
Number of Responses
24
22
20
18
16
14
12
10
All
State
Local
8
6
4
2
0
Yes
No
N/A
Responses
Figure 10. Performance objectives and storm characteristics
Non-storm events, such as the treatment of black ice and bridge frost can be problematic for the
agency. As Figure 11 shows, the majority of the respondents (66%) indicated that performance
measurement was not used specifically for these non-storm events, although several agencies
incorporated such events into the overall snow and ice control plan, and that treating bridges,
black ice, blowing snow, etc. are part of the agency’s snow and ice control procedures. However,
many of these non-storm events are treated after the treatments on the priority routes are
completed. In Sweden, for example, after the precipitation has ended, all Class 1 roads are to be
snow- and ice-free and achieve a 0.25 friction coefficient. Alternatively, the Iowa DOT indicated
that bridge frost events are handled separately and are made part of the forecasting service.
Garages are asked to report on bridge frost, regardless of whether a frost event was forecast.
36
Number of Responses
Performance Measures and Non Storm Events
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Yes
No
N/A
Responses
Figure 11. Agencies using performance measurement used for non-storm events
As Figure 12 shows, the majority of respondents (74%) indicated that they did not use a storm
severity index as a means to classify or rank winter storms so that improved methods can be
developed to combat the storms.
37
Number of Responses
Storm Severity Index
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Yes
No
N/A
Responses
Figure 12. Agencies using storm severity index used
The agencies using an index stated that the indices are still fairly new and under development.
Two respondents indicated using the weather index developed by under SHRP H-350. However,
other agencies have been developing indices to meet their particular needs. The Indiana DOT,
for example, is using a weather index developed by Purdue University that incorporates data
from the National Weather Service and four regions in Indiana. The Iowa DOT reported that it is
working on an index that includes data from its daily reports to provide more detailed weather
information than is available from other sources.
The majority of the respondents (63%) indicated that they do not report road and pavement
conditions to the general public based on any performance measurement system (see Figure 13).
The most common methods used to report road and pavement conditions to the general public
are on the agency web sites and broadcast media, such as radio and television. The use of these
methods allows agencies to reach as many people as possible. However, Sweden reported using
cell phones to report road and pavement conditions to specific individuals
38
Reporting Road and Pavement Condition
Number of Responses
14
12
10
All
State
Local
8
6
4
2
0
n
Dy
am
ic
M
e
a
ss
ge
gn
Si
Co
m
s
m
l
c ia
er
Ra
V
/T
d io
51
1
In
t
n
er
et
w
s
eb
ite
Ot
he
r
Methods Used
Figure 13. Methods to report road condition
Budget and Operations
Virtually every agency reported budget concerns such that budget constraints could impact
performance measurement. Most of the respondents indicated using a type of financial
management system to track expenditures. Several respondents indicated that they budget for a
“normal” winter season, and then track expenditures. Most of the respondents indicated that they
review their activities at the end of the winter season in an effort to gauge performance. Several
agencies indicated that they tracked costs by activity, lane-miles plowed, material usage,
personnel hours, equipment used, and general costs of operations. Budgets are set based on
historical data and respondents indicated availability of budget information dating back years
that they can use to track expenditures. For example, Maryland has a Quality Assurance Team
that prepares a detailed report following the season to review its performance and that of the
weather service provider and the Colorado DOT uses the survey and expenditure information to
form its LOS-based budget. In spite of the detailed cost information little information is
available on determining effectiveness.
As Figure 14 indicates, 58% of the respondents that measure performance segment the highways
for measurement. Most of the roadways were segmented by location, either by maintenance
district, county, or highway route. Other areas were segmented based on traffic volume. A
unique method of segmenting the measurement area is used in Colorado, which uses a random
number generator to select mileposts within a number of routes and then classifies the selection
39
by ADT and roadway type. This segmentation is done to prioritize resources to where they are
most needed.
Number of Responses
Segmenting Highway Areas
24
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Yes
No
N/A
Responses
Figure 14. Agencies segmenting highway areas
Benefits of Performance Measurement
Figure 15 shows that slightly more than one-third (34%) of the respondents indicated that
performance measurement led to improved decision making in snow and ice control. Improved
communication, both internally and externally, were listed as important benefits as well. There
were only slight differences in the responses from state or local agencies. Both groups identified
similar benefits. Other described benefits included uniformity of services delivered, more
effectiveness in products being applied, and the ability to present a budget model that supports
funding levels necessary to achieve the target LOS. These benefits can lead to a more uniform
LOS throughout the state while reducing expenditures.
40
Number of Responses
What benefits does agency obtain from
performance measurement
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
All
State
Local
Improved business
practices with
contractors
Improved
communications
with staff
Improved
decicision
processes relating
to snow and ice
control
Improved external
communications
Other
Benefits Identifed
Figure 15. Agencies identifying benefits
Technology Applications
The use of automated vehicle location (AVL) and global positioning systems (GPS) on
equipment was a popular choice among the responding agencies to help better measure
performance. Contractors in Ontario, for example, use AVL and GPS, Saskatchewan also uses
GPS to track its trucks and materials and the New York State DOT has vehicles equipped with
AVL. The AVL systems allow agencies to track equipment and better determine where resources
are to be deployed.
The deployment of friction wheels is another technology application used by some of the
respondents. This is a technology that has had mixed results. While these devices have been used
extensively in the aviation industry to measure runway friction, they have not been widely used
on highways. The Ohio DOT is currently evaluating the use of friction meters. Sweden has
incorporated friction measurement into its performance standards. Friction coefficients have
been established for each class of roadway that are to be met by the contractors. Conversely, the
Iowa DOT experimented with friction wheels, found the concept to be sound, but the devices to
be unreliable and too costly to deploy statewide. Ontario is considering a pilot project using
friction measurement. Previous testing has shown it to be promising.
Other technologies being used include expanding the uses of RWIS data in making operational
decisions. Road Weather Information Systems (RWIS sites) are typically owned by
41
transportation agencies and automatically report ground (or roadway) level weather variables
such as ground, pavement, and bridge deck temperatures, wind speed at the surface, the presents
of surface level participation, pavement contaminates, and other ground level weather
parameters. For example, the Maryland State Highway Administration uses data from RWIS
stations to supplement and occasionally validate that the pavement is actually bare as reported.
RWIS data are also used in some areas to verify the salinity of the road treatment. The Iowa
DOT uses automated traffic recorder (ATR) data in some areas to obtain traffic speed, and plans
to experiment with combining speed and RWIS data in its performance measures program.
The New York State DOT, New Mexico DOT, and Nevada DOT reported that RWIS data are
unreliable or that the coverage was lacking. Conversely, the Iowa DOT and Maryland State
Highway Administration, report expanding the use of RWIS and combining those data with
other applications. At present, the use of RWIS is widespread throughout the country and is
critical to winter maintenance operations. The Federal Highway Administration project on
environmental sensor stations will produce consistent guidance for state and local agency
personnel responsible for procuring, situating, operating, and maintaining environmental sensor
stations along the roadways.
Another challenge that maintenance officials face is the reliability of weather forecasts. Three of
the respondents indicated the need for better weather forecasting, specifically improved timing
of the forecasts as to when the events start and stop, and the types and measurements of
precipitation forecasted.
Several respondents indicated the need for instrumented means of receiving real-time road
condition data, e.g. friction levels, plow status, and pavement condition. At present, much of the
road condition information received is through visual means, usually from field personnel.
Instrumented maintenance vehicles have been tested in several states over the past few years. For
State Transportation Agencies in Iowa, Minnesota, and California have evaluated a range of
technologies on snow plows that have had varying degrees of success. Many of the concepts
tested during these evaluations have been widely accepted. For example, pavement sensors,
AVL, and computerized material applicators have found acceptance, while technologies such as
heads-up displays, friction meters, and salinity testers are not as widely accepted, and some of
these devices are currently being tested.
The overall majority of the responses stated that improved technology, workforce training and
education, and better proactive maintenance practices helped improve performance. The
improved technologies cited by the respondents included improved trucks, sprayers, and
communications. Improved training with the workforce related to changing practices to be more
proactive with the approaching storm, for example, pre-wetting and anti-icing efforts and
treating roadways prior to a storm. The responses indicate a willingness of the agencies to try
new and innovative technologies. Many of the respondents share ideas and resources through
associations such as the Aurora program, APWA, Clear Roads, and the Pacific Northwest
Snowfighters. All of these improvements listed by the respondents are improvements in
productivity to gain more ground and getting the most out of the resources available to them.
42
The majority of the responding agencies reported budget personnel constraints and resistance to
change from either staff or management as barriers to improving performance with almost onethird (31%) of the responding agencies indicating budget concerns as a major barrier. Clearly,
without proper resources, the agencies will have difficulty experimenting with or funding new
technologies. These are issues felt by most managers, in that there are increasing customer
expectations for snow and ice control and more roads to plow at all levels, but the budgets are
not keeping up.
3.2. Summary and Conclusions
The survey was sent to 162 agencies in the United States, Canadian provinces, Sweden, and
Japan. A total of 43 agencies responded to the survey. The survey indicates all levels of
measuring performance from no performance measurement all to those that incorporate
performance measures into their management plans. There are also indications for improved
methods to measure performance for snow and ice control through technology. Clearly there is
room for improvement in this area.
Most of performance measures cited by the respondents are tied to accounting and management
systems, including lane-miles plowed, personnel hours, overtime hours, tons of material used,
amount of equipment deployed, and cost of operations. Other measures used include time to bare
pavement, time to return to a reasonably near-normal condition, LOS, and customer satisfaction.
Customer satisfaction was cited by 21 respondents as a performance measure. Additionally, 19
respondents indicated that public was surveyed periodically, either by the department or in a
city-wide survey and that the public was generally satisfied with the performance. Two
respondents indicated that they measured customer satisfaction based on telephone calls or
complaints.
The majority of the measures critical to the respondents’ snow and ice control operations focused
on public safety and mobility. Obviously, these subjects are central to the role of all
transportation agencies, so it makes sense that the performance measures would focus on these
subjects. By maintaining mobility and traffic flow, accidents are reduced and public safety is
enhanced.
While state and local agencies are generally interested in providing the best service to the public,
budget and staffing constraints make it difficult for agencies to experiment with new methods or
technologies. Because agencies want to be able to provide these services at the lowest possible
costs, performance measures that are established cannot be too time consuming or costly to
measure.
Eventually, more winter maintenance agencies will adopt more performance measurement
practices. The public will continue to expect clear roads and less harm to the environment from
snow ice control operations. Technologies such as AVL, GPS, friction meters, and RWIS will
help obtain additional data to enhance measuring performance. Expanded use of these
technologies will lead to reduced prices as production and competition increases. Both field
43
personnel and management would have to focus more on outcomes when using these more costly
technologies.
The objectives selected by each agency can drive performance measurement by creating targets
toward which activities can be directed. Performance measures need to also include a short-term
result, an improvement strategy, hold entities accountable and responsive data systems so that
accurate and timely data are generated.
In general, performance measurement for snow and ice maintenance is very mixed bag of
measures used by specific agencies throughout the world. They vary by time interval; continuous
data, storm-by-storm data, and seasonal measurements; they vary by focus; input, output, or
outcome; they vary by data collection technology and archiving methodology; to collect data
some use visual subjective collection methods, and others use repeatable objective collection or
automated collection; and they vary by the degree each agency holds managers accountable to
meet performance goals. Before choosing from the rich variety of performance measures, an
agency must understand its goal and objectives and it must understand the cost implication of
collecting performance measures. For example, one state agency collects bare pavement regain
time from each operator, each operator is trained so that subjectivity is removed from the
measurement, and the data are archived in a statewide GIS for later management review.
Another state transportation agency has district level staff annually select 14 knowledgeable
highway systems users (trucking company managers, ordinary commuters, transit managers,
etc.) to serve on an annual winter maintenance evaluation panel. Every week, about ten
individuals from each district’s panel are called and asked their impression of the snow and ice
removal services on state owned roads. These opinions are compiled and reported as
performance measures. These systems are very different, have very different cost implications,
and probably offer very different outcomes. However, both are measure performance, both are a
valuable tools, and each method achieves an agency goal and objectives.
CHAPTER 4. SELECTED PRACTICES FOR PERFORMANCE MEASUREMENT
During the process of reviewing literature and surveying transportation agencies, considerable
insight was gained into the operations and use of performance measures by several agencies. An
overview of how these approaches and measures used by agencies are described. The agencies
and practices that are included here are those that reported methods to help them save time,
reduce labor, cut costs, increase their level of service, or otherwise improve their ability to get
the job done.
4.1. State Transportation Agencies in the United States
Alaska Department of Transportation and Public Facilities
The survey revealed that the Alaska Department of Transportation and Public Facilities (Alaska
DOT and PF) utilizes and evaluates the performance of both internal department personnel and
contractors. It also links winter maintenance performance to contractor payment. The Alaska
DOT and PF uses customer satisfaction as the principal measure of overall performance of snow
44
and ice control operations. Although no specific metric to gauge satisfaction was noted,
satisfaction was measured through a customer survey at the end of each winter season, and
annual surveys have shown that the public has been highly satisfied with the agency’s efforts.
The Alaska DOT and PF noted that the time to complete clean up after a storm event in urban
areas is one of the most critical performance measures. and that travel speed was of some
importance but was not definitively quantified by the department. The time to clean up after a
storm event in urban areas is measured after each event, based on a visual inspection by
maintenance personnel, and subsequently averaged for an entire season. An annual average over
the entire winter season of 18 hours was established for satisfactory performance. Performance
measures or levels are not set for non-urban areas. The Alaska DOT and PF dedicates more
effort to higher classified roadways.
The department identifies and tracks all snow and ice activities that are performed with
objectives of reducing highway fatalities, achieving customer satisfaction, and keeping traffic
moving safely. External communications was identified as a principal means of measuring
performance, although automated cycle time and amount of materials being expended are some
of the desired information the agency that are not readily available. Additional resources, new
technology, and a dedicated staff comprise the most important factors that contributed to the
agency’s improved performance in recent years.
California Department of Transportation
Information provided by the California Department of Transportation (Caltrans) in responses
indicated that Caltrans utilizes internal staff to perform snow and ice control and measures
several performance indicators covering inputs, outputs, and outcomes such as:
•
•
•
•
•
•
•
•
•
•
•
•
Time to bare pavement
Time to return to reasonably near-normal winter conditions
Traffic flow/LOS
Customer satisfaction
Crash rates
Traffic volumes during storm event
Time for traffic volume to return to normal after storm
Lane-miles plowed
Personnel hours and overtime hours
Tons of materials used
Amount of equipment deployed
Cost of winter operations per lane-mile
In addition, total time of road closures, hours of chain restrictions, accuracy of weather forecasts,
timing and amount of snow received, accuracy of travel services (such as changeable message
signs, radio advisories, chain control signs, etc.), and number of people assigned to snow duty
during storm (in addition to personnel hours) are also considered. The above items were chosen
to help improve safety and mobility across the state. The percentage of time a route is closed
45
during a storm event, percentage of time under chain restrictions during a storm, and weather
forecast accuracy are the three most critical winter maintenance measures used. Caltrans.
However, explicit performance levels and standard approaches have not been established for the
first two measures, although predicted weather forecasts are compared to actual outcomes.
Based on annual assessments the targets for performance objectives are met
Although specific performance targets/levels are not set, performance is evaluated through a
variety of ways, including accounting records, visual inspection by maintenance personnel,
reports from the field, closed circuit television cameras, customer surveys, and the department’s
accounting program for tracking road maintenance activities and associated costs, (the Integrated
Maintenance Management System.)
The department segments its routes into “snow-affected lane-miles,” which are determined by
the elevation of snowfall and may vary by storm. Overall, Caltrans feels that utilizing
performance measurements, results in improved communications with staff, improved decision
making and performance, and improved external communications. The need for more accurate
measurement of precipitation type and amount as well as storm start and end times was noted
and that reduced funding and personnel was regarded as the most significant barriers to
improving performance. Caltrans currently surveys the public concerning snow and ice control
performance every two years, using the Internet; results indicated that the department is
performing to the public’s satisfaction.
Colorado Department of Transportation
The Colorado Department of Transportation (CDOT) utilizes internal department staff and local
government agencies in some instances, to perform snow and ice control operations. CDOT
evaluates performance of the internal and contracted governmental staff and measures
performance of snow and ice control, using the following measures:
•
•
•
•
•
•
•
•
•
•
Time to bare pavement
Traffic flow or LOS
Customer satisfaction
Lane-miles plowed
Personnel hours
Overtime hours
Tons of material used
Amount of equipment deployed
Miles traveled with plow down
Cost of operations per lane-mile
In addition, the number of road closures and duration of each, number of chain events and
duration of each, percent of maintenance employees completing required courses, percent of pretrip and post-trip reviews done on a fleet, percent of equipment operable at beginning of storm,
46
and percent of materials available for snow event are considered. The time to bare pavement is
considered the most critical measure the department uses.
CDOT has a written procedure establishing service levels for different road classes, based on
ADT, but no standard is available to judge whether this goal is met in a timely manner. Several
approaches to collecting data, including accounting records and random visual inspection, are
used as input into CDOT’s mature maintenance management system that tracks snow and ice
control costs. The department also conducts periodic customer surveys to collect data on
performance.
Performance targets or objectives vary by storm characteristics, which includes non-storm
events, as well as by road classification. Maintenance personnel are required to achieve bare
pavement faster on roads with higher traffic volumes than those with lower volumes, using a
method that tracks both factors. CDOT also provides information on road conditions to the
public through the Internet, commercial radio and television, and dynamic message signs. CDOT
reported that efforts into measuring performance have resulted in improved business practices,
improved internal and external communications, improved decision process, and support of a
budget model that subsequently supports necessary funding levels for achieving target service
levels.
CDOT reported that improvements in performance are attributed to chemical de-icers and antiicers and improved communications with the public, and noted that consistency in reporting of
weather and road conditions is the most significant barrier.
Indiana Department of Transportation
The Indiana Department of Transportation (INDOT) utilizes only internal staff to perform snow
and ice control operations. The department is in the process of developing measures that will
ultimately be based on several factors, such as, customer/public concerns, desired LOS, and the
ability to evaluate or “measure” a specific measure.
The department uses an annual winter storm severity index developed by Purdue University that
utilizes data from the National Weather Service and four regions within Indiana and currently
tracks material usage monthly and maintains ten-year averages. The department sees a need for
information on actual road conditions during and after a storm event.
INDOT’s ability to improve performance in recent years is attributable to management support
for developing and supporting a winter operations team that has shared and tried new concepts,
communication of ideas and experience with other states, and participation in research groups
and initiatives. Funding and resistance to change were noted as the most significant barriers to
improvement of winter maintenance performance.
47
Iowa Department of Transportation
The Iowa DOT utilizes and subsequently evaluates the performance of both internal department
personnel and other governmental agencies in winter maintenance. The Iowa DOT measures
several performance indicators including the following:
•
•
•
•
•
•
•
•
Time to return to reasonably near-normal winter conditions
Time to provide one wheel track
Traffic flow
Lane-miles plowed
Personnel hours
Overtime hours
Tons of material used
Costs per lane-mile
The Iowa DOT tracks material, labor, and equipment hours and costs. The department has
utilized several items from winter supervisor daily reports and weather information (from RWIS,
automated weather observing systems, etc.) and continues using other items, such as friction and
crash data, as potential indicators of performance. In recent years, the Iowa DOT has also been
evaluating speed data from existing ATRs as a performance measure. The response noted that
the department has also used surveys of customers, as well as the state patrol, in the past.
Accounting records and field reports are the primary sources of performance data used in Iowa.
The Iowa DOT considers any measure with a direct impact to travelers, such as speed, volume,
or crashes to be critical to its operations. Current objectives are safety, returning roads to nearnormal driving conditions as soon as possible, and using the right type and amount of deicing
materials at the right place and time. The department also acknowledges the need to achieve a
balance between budget, customer service, and the environment. Iowa is one of the states using
different performance levels or targets for different roadway classifications. The Iowa DOT also
measures its performance during non-storm events and has begun to utilize a weather severity
index.
The department segments the road network by class and garage area. Measuring performance
and setting quarterly and annual targets were reported to have improved the decision process for
snow and ice control operations. In recent years, Iowa has tried various approaches and new
technologies to measure performance, found friction wheels to be too expensive and is now
focusing on correlating speed data with weather data to measure snow and ice control
performance and is also working on the use of crash data. The weather index is being used to
determine performance by linking it with budget records.
In order to improve its overall measurement of performance, the agency noted the need for
improving speed data and subsequent impact on traffic operations and potential delay. The Iowa
DOT surveys the public about snow and ice control performance and found that expectations are
higher than originally thought. Furthermore, rating snow and ice removal is a top priority, and
the public has indicated that the Iowa DOT is doing a good job. The department attributes this
48
success to proactive operations, materials used, and improved equipment. Driver behavior,
budget, and staffing were considered the most significant barriers to success.
Kansas Department of Transportation
The Kansas Department of Transportation (KDOT) utilizes only internal staff to perform snow
and ice control operations. KDOT measures winter maintenance performance, using a “level of
service” based on road condition across road categories. KDOT’s objectives were noted as
providing a safe travel way and using resources efficiently. The specific measures used by
KDOT to indicate a safe travel way vary across three roadway categories, but not across storm
type or characteristics. Data indicating these measures are derived from reports by field
personnel and a computer system where field personnel record road conditions. The three
categories, measures, and performance levels are as follows:
•
•
•
Category I. Two bare/wet wheel paths
Category II. Both lanes on two-lane roads with intermittent bare/wet wheel paths
Category III. One wheel path on two-lane roads with intermittent bare/wet wheel
paths
KDOT utilizes a storm severity index, adopting the winter index from SHRP H-350 and utilizing
data from the National Weather Service. Road condition information is reported to the public in
Kansas via the 511 system and an Internet web site.
Massachusetts Highway Department
The Massachusetts Highway Department relies heavily on hired equipment for snow and ice
operations. Currently, the department is entering the second year of a two-year contract with the
hired equipment vendors, and, is considering options to provide a fuel adjustment to the vendors
because of the increasing cost of fuel. The department is currently considering the following:
•
•
•
•
The adjustment will be in the form of an additional payment (in addition to the contract
hourly rate).
The adjustment based on fuel price data available on the Internet from the U.S.
Department of Energy (via the Energy Information Administration).
The adjustment based on the assumed fuel consumption for the equipment, using a factor
that takes into account the miles traveled in an hour’s worth of work, engine efficiency
(miles per gallons), fuel consumption during idling, and the an assumed split between
idle time and running time (each hour).
The fuel adjustment considering amount of the fuel price increase (starting from the
beginning of the contact) multiplied by the assumed consumption rate (gallons/hour).
In preliminary work, the department has been using a typical engine efficiency of six miles per
gallon (equipment under load), and an idling fuel consumption rate of three-quarters of a gallon
per hour. The overall fuel consumption rate is estimated at approximately five gallons per hour.
49
Minnesota Department of Transportation
The Minnesota Department of Transportation (Mn/DOT) uses bare pavement regain time as a
primary performance measurement. Based on predefined road classes, the following statewide
range in hours is determined for bare pavement. (Keranen 2002):
Table 4. Minnesota DOT Pavement Regain Time by Roadway Class
Road class
SC: Super commuter (>30,000
AADT)
U: Urban commuter (10,000–30,000)
R: Rural commuter (2,000–10,000)
P: Primary (800–2,000)
Statewide range (hrs.)
SC1 to SC10
U1 to U10
R1 to R10
P1 to P10
S: Secondary (<800 )
S1 to S10
Mn/DOT has developed a graphical representation to indicate the performance on a particular
road segment, called the Bare Lane Indicator. Graphs are produced that look like automobile
dashboard gauges with green, yellow, and red areas. An arrow is used to show the range of the
response time to bare pavement within the target range. The graphs can show areas of
improvement or enhancement for training, method, funding, equipment, or personnel (Bourdon
2001). Mn/DOT assigned a bare pavement indicator to three road classes, with specific target
values: Super commuter at 1.5 to 2 hours, Urban Commuter at 2 to 3 hours, and Rural Commuter
at 4 hours (Keranen 2002). Other information is collected along with bare pavement regain time
to develop an overall maintenance management plan. Other information gathered includes RWIS
information, salt and sand use, costs per lane-mile, and best practices 48 hours after a storm
event. RWIS information can be correlated with salt and sand use, labor and equipment costs to
analyze current maintenance performance (Keranen 2002). Because precipitation in storm events
varies, specific routes and areas can be analyzed by performance independently from other areas.
New Hampshire Department of Transportation
The New Hampshire Department of Transportation (NHDOT) exclusively utilizes internal staff
to perform snow and ice control operations. The department’s three most important objectives
are uniform compliance with the department’s “Winter Maintenance Snow Removal and Ice
Control Policy”, implementation of salt management plans, and implementation of anti-icing
procedures
NHDOT uses a computerized maintenance activity system that tracks budgets and summarizes
the cost of snow and ice control activities. Improved weather forecasting, improved equipment,
and increased training of employees have been the three most significant factors contributing to
the department’s improved overall efficiency in the past few years. Funding for new
technologies was noted as the most significant barrier to further improving performance.
50
South Dakota Department of Transportation
The South Dakota Department of Transportation (SDDOT) utilizes internal staff, private
contractors, and other governmental agencies to perform snow and ice control operations. The
SDDOT does not formally measure performance, although the department does “informally”
consider measures such as time to reasonably near-normal winter condition, traffic flow, and
customer satisfaction. The SDDOT attempts to budget labor- hours, equipment hours, and
materials for a “normal” winter and tracks these items through payroll and inventory systems.
To better manage snow and ice control operations, the department would like to have better
information on the amount and effectiveness of chemicals. SDDOT surveys the public and has
received generally positive responses. Better equipment and experimentation with new methods
and materials are reported to have helped the department improve performance in the past few
years.
Virginia Department of Transportation
The Virginia Department of Transportation (VDOT) uses outcome-based pay when contracting
out winter maintenance operations. Under this system, the contractor receives a pre-agreed upon
lump sum payment for maintaining a given section of road or facility (Bourdon 2001). VDOT
developed an Asset Management Best Practices Manual for snow and ice control operations that
includes a table describing the LOS for snow and ice control. Roads were divided into priorities
1, 2, 3, and 4. The routes are then were given specific treatments with specified outcomes based
on total accumulation. Priority 1 routes are to be treated, plowed and cleared to obtain 100% bare
pavement within a specified number of hours. Priority 2 routes are to receive chemical treatment
and plowing during the storm, with an end result of completion within a specific number of
hours after the end of a storm. Priority 3, residential streets, are to be sanded as needed and
plowed when feasible to provide a passable roadway. Priority 4 roads are to be sanded as needed
and plowed when feasible with no specific end results (VDOT 2005). Variable message signs are
deployed throughout the state for winter maintenance. The contractor must achieve bare
pavement within 24 hours after the end of a storm event.
VDOT has developed a best practices manual that contains performance targets for all activities.
It is just beginning the process of monitoring operations, with the focus on activities that can be
easily measured. VDOT is the first agency to post performance measures and actual performance
on the web (http://dashboard.virginiadot.org/). On this site, an operations section shows a realtime map of system incidents, other sections show success (or failure) to meet budget
projections.
Washington State Department of Transportation
In 1996 the Washington State Department of Transportation (WSDOT) implemented the
Maintenance Accountability Process (MAP) that combines performance measures into an end
result of a performance based service levels. Performance measures are based on customer
51
oriented outcomes or the results that highway users are able to identify once the results are
collected by field evaluations of highway conditions (Baroga 2004).
A pilot project to assess the service levels for snow and ice control activities by field
measurements and performance measures was implemented. The results were assessed in the
spring of 2000 using roadway traction provided at the time of a field measurement and the time
taken to regain bare pavement at the end of a snowfall event as performance measures. The
roadway traction provided was measured once a week and given a rating from 1 to 5. With a
rating of 1 designates bare or completely sanded pavement, and variations from this condition
are given higher point values. To relate the performance to individual storm events, the time to
regain bare pavement after a winter precipitation event was measured in hours and used a second
performance measure. Point values were assigned to different hour thresholds. A value of 1 was
assigned to a road with a fast regain time, increasing to a 5 for a slow regain time. Highway
categories were also taken into account. For example, a highway with a high ADT would need to
be kept bare throughout a storm event to receive a high rating, while a highway with a low ADT
would require a high rating if bare pavement was regained within six hours. In the end, the point
measures were translated into letter grades similar to the LOS ratings: A, B, C, D, or F (Baroga
2002).
WSDOT has also developed a frost index that relates directly to performance measurements in
winter maintenance operations. The frost index can be used to justify the snow and ice budget
overruns and support requests for supplemental funding (McCullouch et al. 2004).
WSDOT uses internal staff for all winter maintenance operations. Performance is measured for
storm and non-storm evens as the time to bare pavement, wet pavement, and the return to
reasonably near-normal winter conditions. The data for the performance management system are
obtained through visual inspection by maintenance personnel and reports from field personnel.
The staff decides if sand or a chemical deicer will be used and a follow up is done to evaluate the
result of the application, considering the following:
For Chemical Applications:
1. Bare Pavement
2. Patches of frost, back ice, slush, or compact snow
3. Wheel tracks bare, frost, snow, or ice encountered regularly
4. 50% of roadway with compact snow and ice buildup
5. Entire roadway covered with compact snow and ice
6. Unable to evaluate current road conditions
For Sand Applications:
1. 100% of roadway has sand present
2. 50% or more of roadway has sand present
3. All emphasis areas have sand present
4. 50% or more of emphasis areas has sand present
5. 50% or less of emphasis areas has sand present
52
6. Unable to evaluate road conditions
The three most important current objectives for snow and ice control operations are to move
towards a statewide chemical priority program, evaluation of all chemical applications to refine
the necessary application in different weather events, and the calibration of all equipment used in
sand and chemical application. These objectives are regularly tracked by application records.
Highways are ranked by priority, from one through five, into treatment categories. Road
conditions are reported to the public based on the performance management system by dynamic
message signs, commercial radio, television, 511, and an internet radio station.
Due to the varying climates within the state of Washington, treatments and performance goals
are divided into east and west regions. As funding levels and other resources require
prioritization of different roads for snow and ice control services, different treatments are
employed for individual roads and sections of roads. Area supervisors choose sand or chemical
deicers applications to meet the goals. If chemicals are used, the time to wet or bare pavement is
measured. If sand is used, a follow-up evaluation is conducted to determine conditions, e.g., how
much of roadway has sand present (100%, 50%, all emphasis areas, etc.)
WSDOT has developed LOS measures based on visual description of conditions supplemented
with pictures. The agency also describes four treatment levels that are linked to LOS. “Levels of
Service” (LOS) are reported on a scale of “A” through “F” are defined as follows:
•
•
•
•
•
LOS A: A very high LOS in which the roadway and associated features are in
excellent condition. All systems are operational and users experience no delays.
LOS B: A high-maintenance service level in which the roadway and associated
features are in good condition. All systems are operational. Users may experience
occasional delays.
LOS C: A medium-maintenance service level in which the roadway and associated
features are in fair condition. Systems may occasionally be inoperable and not
available to users. Short-term delays may be experienced when repairs are being
made, but would not be excessive.
LOS D: A low-maintenance service level in which the roadway and associated
features are kept in generally poor condition. System failures occur because it is
impossible to react in a timely manner to all problems. Occasionally, delays may be
significant.
LOS F: A very low service level in which the roadway and associated features are
kept in poor and failing condition. A backlog of system failures would occur because
it is impossible to react in a timely manner to all problems. Significant delays occur
on a regular basis.
The department’s efforts are prioritized by service levels. High-priority service levels are
directed to major highways (such as Interstate 90), and other highways are assigned appropriate
service levels. These service levels range from Level 1 to Level 4. The service levels represent
the expected condition after the treatments are completed and the storm event is ended. On a
Level 1 roadway, the department attempts to make the roadway bare and dry or bare and wet as
soon as possible. Level 2 roadways may have snow and ice buildup at times. Level 3 roadways
53
can have snow and ice buildup on a regular basis, and Level 4 sections are often covered with
compact snow. Each service level has a corresponding roadway treatment action using liquid
anti-ice chemicals, solid chemical treatment, plows, and sand.
Wisconsin Department of Transportation
The Wisconsin DOT uses a single measure to measure the LOS for snow and ice operations.
Periodic field condition surveys are conducted to measure the traction conditions on the travel
lane of the road surface. These conditions are determined by observing a mile of road after a
winter operation has taken place. Bare pavement is considered if 95% of the roadway is free
from ice and snow. A roadway is considered sanded if at least 60% of the travel lane has sand on
the surface. This is equal to a travel lane with bare tire tracks with sand on the remainder of the
lane (Conger 2005). The state reimburses 72 counties to perform winter maintenance on state
and federal roads (Adams et al. 2003).
The Wisconsin DOT’s performance measurement program has been adapted from that of the
WSDOT. Winter maintenance operations is contracted out to county highway departments, and
performance is measured as the time to wet pavement, by customer satisfaction, crashes per
VMT, cost of winter operations per lane-mile, and percent of salt spreaders/controllers
calibrated. The most critical measures are public satisfaction and time to bare/wet pavement;
however, target measures have not yet been established. The three most important current
objectives are providing bare/wet pavement in a reasonable amount of time and effort, improving
the coefficient of friction between vehicle tires and the pavement, and providing good winter
driving conditions using the most efficient methods possible. The methods used for obtaining
data for the performance measurement system include accounting records, visual inspection by
law enforcement, and periodic customer surveys.
Ontario Ministry of Transportation
In order to develop performance measurements and benchmark methodology for Ontario’s
municipal roads, the Ontario Good Roads Association formed a committee of road professionals
in 1997. This committee created an activity map divided into six categories, including winter
control. This map has been adopted by the Ministry of Municipal Affairs for their Municipal
Performance Measurement Program and by the Ontario Municipal CAO’s Benchmarking
Initiative. The committee enlisted a select group of municipalities to use various performance
measures on high traffic volume roads. One winter maintenance group focused on rural arterial
systems, while the other focused on urban local residential systems. The following performance
measures were documented (Anderson 2004):
•
•
•
•
•
•
Cost per lane-kilometer
Annual cm of snowfall
Total annual tons of abrasive, including salt per system kilometer
Total annual tons of salt per system kilometer
Usage of the road system (vehicle-kilometer /lane-kilometer)
Median operating costs per lane-kilometer
54
•
•
•
•
•
•
Average number of winter event responses
Average usage in vehicle kilometer/lane- kilometer
Average percent of plows/salters/combination units, municipally owned
Percent of municipalities pre-wetting salt prior to application to the road surface
Average length of plow route
Percent of municipalities using a wingman in the truck
Non-event response activities such as snow fence operations, winter standby staff and
contractors, winter patrol, winter drainage, spring clean up, and overhead were also documented.
The Ontario Ministry of Transportation measures the performance of private contractors, who
perform all winter maintenance activities, by time to bare pavement. The time to bare pavement,
in hours, varies by road classification and storm characteristics. Road conditions are monitored
by a patrol vehicle and reported by either highway number or patrol number. The data for the
performance measurement system is obtained through accounting records, visual inspection by
maintenance personnel, reports from field personnel and AVL at certain locations. The overall,
most important, objectives are to maintain safe conditions during a storm, recover bare pavement
after a storm event, and minimize salt loading to the environment. In addition to measuring the
time to bare pavement, the number of plows and spreaders operating and response time are
monitored to ensure conformance with established operating guidelines and public safety.
Information, such as daily hours of operation for each piece of contracted equipment and tons of
salt and sand applied, are collected for audit and payment of private contractors.
The Ontario Ministry of Transportation has a system of “Best Practices” and “Levels of
Service.” The “Best Practices” (formerly referred to as “quality standards”) specify how and
when an operation (plowing, salting, sanding) is performed. “Levels of Service” define the
expected end result (level packed snow, centre-bare or fully bare pavement) and the maximum
elapsed time after the storm until that result is achieved. The Ministry also has targets for how
often those service levels are achieved (e.g., 98% of the storms in a winter).
Finland
Road officials in Finland use a patrol vehicle to measure the friction on a roadway and the
operator determines whether the roadway meets the frictional requirements and recall the
maintenance fleet to treat the unsatisfactory location (Al-Qadi et al. 2002).
The Finnish National Road Administration (FnRA) sets the policies and LOS that the contractors
have to meet. The FnRA also specifies the environmental parameters that are to be met. For
example, contractors are required to have the proper knowledge and skill in the use of road salt
so as little salt as possible is used while keeping the road in safe condition.
The road network is divided into five main maintenance classes (Is, I, Ib, II, III) and class Ib has
a corresponding maintenance class, T-Ib, for built-up areas. Each class has a different LOS and
quality standards. In deciding the maintenance class of a road, not only are the classification
55
criteria taken into consideration, but also local conditions, the nature and composition of traffic,
the speed limit, and qualitative integration with the LOS of the municipality’s road network.
Road classes are defined in a logical pattern from the viewpoint of road maintenance personnel.
Thus, snow and ice control operations can be implemented as economically as possible. The road
network is defined in Table 5.
Table 5. Finnish Road Network (Finnish Road Maintenance 2001)
Finland is also a leader in using friction measurements as an indicator of effective snow and ice
control. The FnRA has established standards in using friction, as shown in Tables 6 and 7.
56
Table 6. Quality standards and friction indicators (Finnish Road Maintenance 2001)
Winter maintenance class
Ib an TIb
II
Spot sanding
0.25
Line treatment
0.20–0.22
Is
Road
surface
below -6°C
0.25
I
Road
surface
below -4°C
0.25
Normal
0.30
0.28
0.25
At night
2200–0500*
0.28
2200–0500*
0.25
2200–0500*
As needed
2200–0600*
As needed
2200–0600*
As needed
2 hr
2 hr
Salt 3 hr
Sand 4 hr
6 hr
line sanding
10 hr
line sanding
Friction
requirement
Cycle time
III
K1
K2
According to traffic demands
After 2200*
K1 by 0500*
K2 by 0600*
2 hr
*Time listed in 24 hour format
Table 7. Friction indicators and driving conditions (Finnish Road Maintenance 2001)
Description
of driving
conditions
0.00–0.14
Bad driving
conditions,
wet ice
Very
slippery
0.15–0.19
Icy
Slippery
Friction value
0.20–0.24
0.25–0.29
Tightly packed Rough, packed
snow
ice and snow
Satisfactory
winter
conditions
(based on
friction value)
Good winter
conditions
(based on
friction value)
0.30–0.44
Bare and
wet
0.45–1.00
Bare and
dry
Not
slippery
Not
slippery
According to the administration’s policy, the following friction indicators must be met criteria:
•
•
•
•
•
•
•
•
The friction requirement must be met on at least half of the surface area of the lane.
The friction requirement for Classes Is and I roads is 0.25 when the temperature of
the road surface is lower than the limit value.
In freezing situations Classes Is and I are treated using preventative salting to prevent
slipperiness or to at least minimize its duration.
At night the friction requirement is 0.28 for Class Is and 0.25 for class I.
In Class Ib the friction requirement is 0.25 in early and late winter.
During stable winter conditions class Ib requires sufficient treatment procedures
when the friction value drops below 0.25. The entire length of the road must be
treated no later than when the friction value is expected to drop below 0.20. On
specified busy Ib roads the entire length of the road must be treated no later than
when the friction value is expected to drop below 0.22.
In maintenance Class TIb (built-up areas), salt is used as necessary only in early and
late winter.
In Classes II and III, sufficient friction needed by traffic is required.
57
•
•
•
In Class II, regular anti-slipping procedures are implemented at problem sites, so
traffic ability is guaranteed in all conditions. The entire length of the road is sanded
during particularly difficult driving conditions.
Anti-icing procedures in Classes II and III are supplemented by coarsening the
surface of packed snow.
In Class III, particular problem sites are spot sanded to keep the road in travel
condition. The entire length of the road is sanded during especially difficult driving
conditions.
Snow removal must adhere to the following criteria (also illustrated in Table 8):
•
•
•
•
•
The maximum snow depth must not be exceeded while it is snowing or during
maintenance procedures thereafter.
Only half as much slush is allowed as snow.
Plowing must be started no later than when half of the maximum amount of snow has
accumulated. This starting threshold is not used at night in classes II, III, and K. In
Class Ib and TIb, the starting threshold at night is 4 cm.
The maximum amounts of snow refer to normal snowfalls. In exceptional snowstorms
(a few times a year), these values may be exceeded.
Snow depth refers to the prevailing situation in the lanes, including snow piled by
traffic.
Table 8. Quality standards for snow removal (Finnish Road Maintenance 2001)
Winter
maintenance
class
Maximum
snow depth
when snowing
Is
I
Ib and TIb
II
III
K1
K2
4 cm
4 cm
4 cm (8 cm
at night)
8 cm (10 cm
at night)
10 cm (10
cm at night)
3 cm (8 cm
at night)
Cycle time,
clean after
snow stops
2.5 hr (slush
2 hr)
3 hr (slush
2.5 hr)
3 hr
4 hr
6 hr
3 hr
4 hr
If snowing
after 22 at
night
Plowed clean within cycle
time
05, or cycle
time
06, or cycle
time
06, or cycle
time
05
06
To obtain the proper LOS, well-timed management, seamless cooperation between different
contracts, safety, and environmental friendliness, the FnRA establishes quality assurance plans,
the entire personnel of the contractor operate according to this plan. The quality plan also
functions as a winter maintenance plan, which identifies the issues that have the most impact on
the work, such as operating routes and resource allocation. Post-winter maintenance control
requires reporting of the following items:
•
•
Costs by maintenance class
Implementation of contracts
58
•
•
•
•
•
Quality and the LOS on the road
Road user feedback
Description of wintertime road safety
Environmental impact information
Description of the prevailing weather
A winter maintenance report based on the above items is compiled within FnRA at the contract,
district, and administrative levels. Costs are reported by maintenance class to help compare and
control the price level of different areas. Implementation of contracts (quality) is reported as the
contract supervisor’s personal evaluation and the number of deviation reports, complaints,
penalties and bonuses.
The LOS and quality on the road is reported by means of quality monitoring based on random
sampling. An annual study of the LOS also provides a general overview of the quality of winter
maintenance and especially its development.
Sweden
The SNRA mandates contracting out for all winter service, and its own forces compete in the
bidding. Different highway classes, based on function and traffic volume, require different
standards. The standards are tied to roadway surface temperature, precipitation, and roadway
appearance in different conditions. Friction is measured by a supervisor using a Corabla friction
tester in a light truck, separate from the production truck, (Harrigan 1999).
Sweden stipulates that its “highest volume road shall be free from snow and ice no later than two
hours after the snow has stopped falling if the road surface temperature is above -8 degrees C (18
degrees F) and that during the period when the snow is actually falling, the depth of snow shall
not exceed 2 cm (0.8 in.) and slush depth shall never be more than 1 cm (0.4 in.)” (Olander
2000).
Sweden also surveys 14,000 people annually from all seven regions to link performance
measures to customer expectations. Questions deal with how the SNRA manages snow,
slipperiness, and slush, along with attitudes towards salt use (Harrigan 1999).
The contractor’s payment is linked to performance measurements, such as time to bare
pavement, time to return to a reasonably near-normal condition, friction, customer satisfaction,
and tons of materials used. The most critical items to snow and ice operations are friction and
snow depth.
The three most important objectives for the agency are friction, snow depth, and time to
“normal” conditions. For the measure of friction, a friction value is used as the performance
level, while snow depth is measured in centimeters, and time is measured in hours. The data for
the performance management system are obtained by accounting records, visual inspection by
maintenance personnel, reports from field personnel, calls from the public, and the contractor’s
59
records. The performance measures on roadway conditions are reported to the public by dynamic
message signs, commercial radio and television, 511, an Internet website, and by cell phone.
Norway
The Norwegian Public Roads Administration measures performance by testing friction using a
friction meter. A required LOS of 0.4 (coefficient of friction) is used throughout the country for
public and private operations (Harrigan 1999). Along with measuring friction, photographs,
activity logs and observations are used to evaluate specific friction improvement methods (AlQadi et al. 2002). The administration also uses thermal mapping to improve service and reduce
costs.
Japan
The Hokkaido Regional Development Bureau developed guidelines for goals, based on ADT
volume and road surface conditions, of recommended LOS in winter conditions. Patrolling
inspectors visually measure the effectiveness of the procedures (Pisano, 2004). The Japan
Highway Public Corporation has published a national maintenance manual. Japan also uses a
neural network to predict friction from various data, including weather, traffic, and pavement
condition (Al-Qadi et al. 2002).
The Japanese Ministry of Land Transport and Infrastructure has developed a national
performance measurement program, of which road maintenance is a part. This is an outcomebased program designed to make the Ministry more efficient and accountable to the public. The
Ministry established 17 performance indicators (Table 9) as targets; although snow and ice
control is not specifically mentioned, it falls under the safety policy theme (Japanese Ministry of
Land Transport and Infrastructure 2004).
60
Table 9. Japanese road management performance plan
Policy Theme
Performance Indicator
Current
Indicator Val.
(FY2002)
Target for
FY2003
Target for
FY2007
(under constr.)
1. Vitality
(restoration of
economic
vitality through
urban renewal
and regional
coordination)
1. Time loss due to traffic congestion
(congestion monitoring zone)
610 million man
hr/yr
590 million man
hr/yr
(2.5% reduction)
about 10% reduction
2. Ratio of electronic toll usage
National
Metropolitan expressway
Hanshin expressway
3. Hours of road work
5%
6%
3%
235 hr/km/yr
70%
85%
85%
About 20%
reduction
15%
59% (access to 39
locations)
72%
15%
20%
15%
225hr/km/yr
(4% reduction)
13% (switchover of
2.1 million vehiclekm/day)
61% (access to 40
locations)
73%
63%
64%
68%
17%
21%
About 50%
7%
8%
15%
118.4 incidents/
100 million
vehicle-km
86%
91%
66%
116 incidents/ 100
million vehicle-km
108 incidents/ 100
million vehicle-km
2. Living
(better quality of
life)
3. Safety
(ensuring secure
and safe life)
4. Environment
(preservation and
creation of
environment)
5. Road
administration
reform
4. Ratio of high standard road usage (targeted
traffic that will be newly switched over to
expressways during the current fiscal year)
5. Ratio of roads with access to hub airports
and ports
6. Ratio of main cities in neighboring regions
that are connected to each other by an
upgraded national road
7. Percentage of people who are able to have a
safe and pleasant drive into the city, the center
of daily life, in under 30 minutes
8. Percentage of barrier-free main roads in the
vicinity of passenger facilities with an average
daily user volume of more than 5,000
9. Percentage of trunk roads in urban areas
without telephone poles
10. Ratio of death and injury due to road
accidents
11. Road structure
Bridge
maintenance ratio
Pavement
12. Percentage of cities that have rescue routes
covering a wide area in the event of disasters
13. Reduction of CO2 emission
14. Ratio of NO2 environmental goal
achievement
Ratio of environmental goal achievement
for suspended particulate matter
15. Achievement rate of required limits on
nighttime noise
16. Level of road user satisfaction
17. Number of hits on homepage
13%
-
68%
77%
87%
93%
Maintain current level
68%
76%
64%
Reduce CO2 emission by transportation
sector to about 250 million tons CO2 by 2010
67%
About 80%
-
About 10%
About 60%
61%
63%
72%
2.6 points
15.46 million
access/year
2.7 points
26 million
access/year
3.0 points
100 million
access/year
The Japan Highway Public Corporation, part of the Japanese Ministry of Land Transport and
Infrastructure, has developed a national maintenance manual and is developing regional
manuals. The Hokkaido Regional Development Bureau also has developed a winter road surface
maintenance guideline. The documents describe recommended LOS during wintertime
conditions. The guideline includes road management goals for various highway facilities defined
by combinations of ADT volume and area type. The management goals are defined in terms of
five classes of road surface conditions. Charts relate the classifications of road surface conditions
to ranges of friction coefficients determined by research. The performance evaluation of winter
maintenance operations in Hokkaido is based on a visual inspection of road surface conditions
by patrolling inspectors.
61
In Japan, private contractors are used in snow and ice operations. The performance measures
indicated are tons of materials used and equipment operation hours. The contractors submit
records of the equipment hours and materials used to the Ministry for payment.
Criteria for the mobilization of snow and ice control staff and vehicles and winter road LOS are
set for each road category according to the amount of snowfall, air temperature, and traffic
volume in each cold, snowy region. Snow and ice control operations, including the plowing of
snow, the application of material and the operation of snow hauling, is based on such criteria and
on LOS. The Hokuriku Regional Bureau of the Ministry of Land, Infrastructure, and
Transportation, for example, sets mobilization criteria for each type of winter maintenance
operation.
City of Sapporo, Japan
The City of Sapporo, Japan, sets LOS using photographs and descriptions of snow and ice
conditions (see Tables 10 and 11). The results of the business owner survey led to measuring
performance in terms of the outcome indicators of maintained road width and friction. Winter
maintenance activities included plowing, road width widening, hauling snow, and anti-freezing
agent applications. For securing effective road width, the input activities included the staff, time,
facilities and equipment specifications. The input indicators were targets of results, the results
and correlated achievement rate. The output was securing effective road width, indicated by the
actual measured passable road width.
The outcome measurement was traffic delay and achievement rate based on the number of days
when the effective road width was secured. For the goal of securing skid resistance, the input
was budget allocation based on the totals of annual snowfall. The output was securing friction
and the measure was measuring the skid resistance. The outcome was the number of days when
friction was secured, measured by the number of traffic accidents. The project was evaluated by
a service effectiveness report, financial report, and efficiency indicators (Yamamoto et al. 2004).
Table 10. Snow and ice operations for Sapporo, Japan
Road category
Major trunk road, trunk road
Collector road A
Collector road B
Residential road
LOS and the target road condition
Level 4 (daytime): Powder snow, wet snow,
slush
Level 3 (daytime): Compacted snow, or wet
snow or ice
Level 3: If possible, compacted snow, or wet
snow or ice
Level 2 (daytime): Ice sheet, or powder snow
over ice
62
Table 11. Winter road standards and LOS for Sapporo, Japan (PIARC 2006)
Standards for
highways and
trunk roads
Standards for
sub-arterial
roads 4 and 5
Level 4
Level 3
Standards for
residential
streets
Level 2
Level 1
4.2. Summary
The case studies summarized in this chapter describe agencies that are using various methods to
ensure acceptable levels of snow and ice control performance. Because statistically sufficiently
extensive quality monitoring of the entire road network is difficult and expensive, agencies are
proceeding deliberately to fully implement performance measures programs while experimenting
with systems incrementally. Existing snow and ice control operations still use traditional
practices such as plowing and material spreading while striving to improve the processes. Road
condition and weather information have become a crucial parts of snow and ice control
operations. Agencies are keenly aware of the costs of gaining efficiencies and are taking
advantage of the increasingly available weather data as much as possible to improve snow and
ice control operations.
Snow and ice control agencies have more weather and road condition information better
equipment, more materials, and are providing higher levels of service to the traveling public than
ever before. Likewise, the expectations of the traveling public have risen along with budget
constraints and environmental concerns, to force snow and ice control agencies to incorporate
improved tactics and operations.
CHAPTER 5. SYNTHESIS AND ASSESSMENT OF PERFORMANCE MEASURES
5.1. Introduction
Based on the review of relevant literature and survey of agencies, more than 20 distinct
performance measures were identified. Agencies used a variety of approaches to collect the data
to calculate the measures. Within this data set, more than 40 combinations of approaches and
measures were identified. This chapter categorizes the various measures as input-, output-, or
outcome-based and summarizes their frequency of use.
Generally, the data for input and output measures come from the agencies’ accounting systems
or maintenance logs. There is not much variation in the approach to acquiring these data.
63
However, it is more difficult to obtain data for outcome measures, since the majority of outcome
measures are based on some form of manual observation. Developing technologies in the
experimental stages can provide solutions to acquiring outcome measure data.
Additionally, any measure used for time-series analysis would benefit from applying a storm
severity index. The various indices are provided here based on the availability of data to
calculate the index and its usefulness in improving understanding of performance or
communicating performance to administrators.
To provide direction for this synthesis and assessment, the study team developed criteria (listed
in section 5.3) for evaluating measures and the associated approaches to acquiring data. These
criteria were applied to eliminate measures or approaches that do not exhibit these desired
characteristics:
1.
2.
3.
4.
5.
6.
Related to controllable facets of performance
Reliable
Understandable
Timely
Consistent
Sensitive to data collection costs
5.2. Performance Measures in Use
Snow and ice control performance measures and efficiency measures are also grouped with
outputs in this classification system. Efficiency is a measure of input divided by output. Outcome
measures are based on an assessment of how well operational goals were met; these measures
include time needed to regain bare pavement and the friction coefficient after treatment. Input
and output measures are usually based on accounting records or operational reports, while
outcome measures require some form of monitoring.
The survey of agencies responsible for snow and ice removal revealed more than 20 distinct
measures in use in the United States and other countries. The responses for each measure,
grouped into input, output, or outcome categories, are listed in Table 12.
64
Table 12. Summary of snow and ice control performance measures by category
Input measures
Fuel usage (4)
Overtime hours (18)
Personnel hours (18)
Percent of salt spreaders calibrated (8)
Output measures
Lane miles plowed (15)
Tons of material used (20)
Amount of equipment deployed (14)
Plow-down miles traveled (4)
Cost per lane mile (efficiency) (15)
Outcome measures
Time to bare pavement (9)
Time to wet pavement (3)
Time to return to a reasonably near-normal winter condition
(10)
Time for traffic volume to return to “normal” after the storm (5)
Time to provide 1 wheel track (1)
Friction (5)
Level of service (11)
Travel Speed during storm (2)
Customer satisfaction (18)
Crashes per vehicle mile (2)
Traffic volume during storm (2)
Table 13 reports the agencies responding to the survey that use various performance measures.
This table is the inverse of a case study, which focuses on a single agency; Table 12 instead
focuses on the measure.
65
Table 13. Agencies using various performance measures
Input Measures
Fuel usage (4)
Personnel hours (18)
New Mexico DOT
Ohio DOT
Saskatchewan Department of Highways
New York State DOT
and Transportation
Caltrans
Washington Co., Minnesota
Detroit
Cuyahoga Co., Ohio
Indianapolis DPW
El Paso Co. Colorado
McHenry Co., Illinois
Overtime hours (18)
New Mexico DOT
Ohio DOT
City of Seattle
New York State DOT
Douglas Co., Nebraska
Caltrans
Edmonton, Alberta, Canada
Detroit
City of West Des Moines, Iowa
Indianapolis DPW
Saskatchewan Department of Highways
El Paso Co. Colorado
and Transportation
McHenry Co., Illinois
Washington Co., Minnesota
New Mexico DOT
Cuyahoga Co., Ohio
City of Seattle
Iowa DOT
Douglas Co., Nebraska
Minnesota DOT
Edmonton, Alberta, Canada
Ada County Idaho
City of West Des Moines, Iowa
Saskatchewan Department of Highways
and Transportation
Percent of salt spreaders calibrated (8)
Washington Co., Minnesota
New York State DOT
Cuyahoga Co., Ohio
El Paso Co. Colorado
Iowa DOT
McHenry Co., Illinois
Minnesota DOT
Edmonton, Alberta, Canada
Ada County, Idaho
City of West Des Moines, Iowa
Saskatchewan Department of Highways
and Transportation
Ada County Idaho
Minnesota DOT
Output Measures
Lane miles plowed (14)
Amount of equipment deployed (14)
Caltrans
Ohio DOT New York State DOT
Ohio DOT
Caltrans
Detroit
Detroit
King County, Washington
Indianapolis
El Paso Co., Colorado
Cedar Rapids, Iowa
McHenry Co., Illinois
McHenry Co., Illinois
New Mexico DOT
City of Seattle
Douglas Co., Nebraska
Douglas Co., Nebraska
Edmonton, Alberta, Canada
Edmonton, Alberta, Canada
City of West Des Moines, Iowa
Saskatchewan Department of Highways
Saskatchewan Department of Highways
and Transportation
and Transportation
Cuyahoga Co., Ohio
66
Washington Co., Minnesota
Iowa DOT
Ada County Idaho
Plow-down miles traveled (4)
New Mexico DOT
New York State DOT
Edmonton, Alberta, Canada
City of West Des Moines, Iowa
Ada County, Idaho
Minnesota DOT
Tons of material used (19)
Sweden
Ohio
New York State DOT
Caltrans
Detroit
Indianapolis
King County, Washington
Cedar Rapids, Iowa
McHenry Co., Illinois
City of Seattle
Douglas Co., Nebraska
Edmonton, Alberta, Canada
Maryland State Highway
Administration
City of West Des Moines, Iowa
Saskatchewan Department of Highways
and Transportation
Washington Co., Minnesota
Cuyahoga Co., Ohio
Iowa DOT
Ada County, Idaho
Cost of winter operation per lane mile
(efficiency) (15)
Ohio DOT
New Mexico DOT
Iowa DOT
Maryland State Highway
Administration
New York State DOT
CalTrans
Detroit
Indianapolis
King County, Washington
El Paso Co. Colorado
Edmonton, Alberta, Canada
City of West Des Moines, Iowa
Saskatchewan Department of Highways
and Transportation
Washington Co., Minnesota
Minnesota DOT
Outcome Measures
Time to bare pavement (10)
Travel speed during storm (2)
Sweden
Ohio DOT
New York State DOT
Ada County Idaho
Caltrans
(Iowa is evaluating but has not
Ontario, Canada
incorporated into operations)
Maryland State Highway
Administration
Customer satisfaction (18)
Edmonton, Alberta, Canada
Sweden
McHenry Co., Illinois
Ohio DOT
Cook County Illinois Highway
New York State DOT
Department
Caltrans
Ada County Idaho
Alaska Dept. of Transportation and
Minnesota DOT
Public Facilities
Detroit Street Maintenance Division
Time to wet pavement (3)
King County, Washington
Missouri
El Paso Co., Colorado
Washington Co., Minnesota
Cedar Rapids, Iowa
Ada County, Idaho
67
McHenry Co., Illinois
New Mexico DOT
Douglas Co., Nebraska
Edmonton, Alberta, Canada
City of West Des Moines, Iowa
Washington Co., Minnesota
Minnesota DOT
Cuyahoga Co., Ohio
Ada County Idaho
Time to return to a reasonably nearnormal winter condition (10)
Sweden
Ohio DOT
New York State DOT
Caltrans
Iowa DOT
Saskatchewan Department of Highways
and Transportation
Friction (5)
Edmonton, Alberta, Canada
Sweden
El Paso Co., Colorado
Ohio DOT
Douglas Co., Nebraska
Douglas Co., Nebraska
City of West Des Moines, Iowa
Cuyahoga Co., Ohio
Ada County Idaho
Time for traffic volume to return to
“normal” after the storm (5)
Crashes per vehicle mile (2)
Caltrans
CalTrans
Detroit Street Maintenance Division
Ohio DOT
Cedar Rapids, Iowa
Cuyahoga Co., Ohio
Ada County Idaho
Traffic volume during storm (2)
CalTrans
Ada County Idaho
Time to provide 1 wheel track (1)
Iowa DOT
Time until low-volume roads open to
traffic
Level of service (11)
Missouri
Ohio DOT
New York State DOT
Caltrans
Annual quality assurance reports
Iowa DOT
Ohio
New Mexico DOT
Detroit Street maintenance Division
Time since last treated
Kansas DOT (Pavement condition for
Indianapolis
the
category of route)
Contract trucks deployed in a
Cedar Rapids, Iowa (traffic flow)
reasonable manner
City of Seattle
Maryland State Highway
Saskatchewan Department of Highways
Administration
and Transportation
Ada County Idaho
Total Hours Road Closed
CalTrans
68
Table 14 lists the approaches used by agencies to acquire data for outcome measures. Clearly
human observation is the most common approach.
Table 14. Outcome measures and approaches used by responding agencies
Measure
1. Time to reasonably nearnormal winter conditions
2. Customer satisfaction
3. Travel speed
4. Time to bare pavement
5. Total time of road closure
6. Total time of chain
restrictions
7. Time to single bare wheel
track
8. Time to two bare wheel paths
9. Time to treat critical areas
10. Friction
Approach
1.1 Visual inspection by maintenance personnel (AK)
1.2 Visual inspection by maintenance personnel (CA)
1.3 Reports from field personnel (IA)
1.4 Reports from field personnel (CA)
1.5 Reports from field personnel (NV)
1.6 Visual inspection by maintenance personnel (NV)
1.7 Reports from field personnel (NM)
1.8 Visual inspection by maintenance personnel (NM)
1.9 Visual Inspection by law enforcement (NM)
1.10 Reports from field personnel (NY)
1.11 Visual inspection by maintenance personnel (NY)
2.1 Annual survey at end of season (AK)
2.2 Internet survey (CA)
3.1 Automatic traffic recorders (NY)
3.2 Testing automatic traffic recorders (IA)
4.1 Visual inspection by maintenance personnel (CO)
4.2 Report from field personnel (CO)
4.3 Visual inspection by maintenance personnel (MD)
4.4 Reports from field personnel (MD)
4.5 Reports from field personnel (MO)
4.6 Reports from field personnel (NV)
4.7 Visual inspection by maintenance personnel (NV)
4.8 Visual inspection by maintenance personnel (OH)
4.9 Reports from field personnel (OH)
4.10 Visual inspection by maintenance personnel (WA)
4.11 Reports from field personnel (WA)
4.12 Visual inspection by law enforcement (WI)
4.13 Visual inspection by maintenance personnel (ON)
5.1 Accounting records of hours closed (CA)
6.1 Records of chain restriction hours (CA)
6.2 Records of chain restriction hours (CO)
7.1 Reports from field personnel (IA)
7.2 Reports from field personnel (KS)
8.1 Reports from field personnel (KS)
9.1 Reports from field personnel (MO)
10.1 Testing (OH)
10.2 Established friction coefficient (Sweden)
10.3 Preliminary testing (ON)
69
Other practices that emerged in the survey comments include the following:
•
•
•
•
•
A PSIC chart to help identify uniform pavement conditions by combining traffic flow
characteristics and visual observation.
Various outcome measures are sometimes combined to form an overall LOS rating
for a roadway.
Contracts with private sector operators are written such that reimbursement is based
on a combination of input (pay items) and output or outcome measures
(expectations).
Innovative technologies such as AVL, GPS, friction meters, and various sensors of
materials, equipment, and temperature are installed on winter maintenance vehicles to
help collect performance measure data.
Winter weather severity indices have been developed to help quantify the relationship
between the severity of winter weather events and roadway condition or safety
factors.
5.3. Screening of Approaches
Analysis of the survey findings identified 4 input measures, 5 output measures, and 11 outcome
measures used by public agencies to measure snow and ice control performance. To identify
measures and approaches that warrant further study, the following criteria were applied to
available measures and approaches:
Measure criteria
•
•
•
•
•
•
Does the measure directly measure safety, mobility, or public satisfaction?
Does the measure improve snow and ice control?
Is the measure mapped to roadway segments?
Is the measure reported for garages or districts?
Is the measure sensitive to storm characteristics?
Does the measure examine storm events individually or annually?
Approach criteria
•
•
•
•
•
Is the approach quantitative?
Is the approach stable across observers?
Is the technology likely to improve?
Is a major capital or operational investment required?
Can the approach be “piggy backed” on another system to reduce installation cost?
Applying these criteria revealed that input and output measures are valuable management tools
because they measure the amount of material, labor, and money consumed, as well as the amount
of material applied to roads, lane-miles plowed, etc. However, these measures do not directly
address the goals of the agencies, regarding public safety and maintenance of mobility. As they
70
are, input and output measures help with budgeting and can be used roughly to compare
efficiency between garages or districts that experience similar snow and traffic conditions. The
measures do not improve snow and ice control, but help track the investment required to do so.
The measures are generally not mapped to roadway segments, although they are often reported
by garage or by district. Input and output measures are not observed to be sensitive to storm
characteristics, although they could be if an index were applied.
Input and output measures rate high on the “approach” criteria because the investment in
accounting and maintenance management personnel is used to produce the reliable and
quantitative numbers. Input and output measures are reasonably easy to obtain and measurement
is stable from year to year. However input and output measures do not measure safety or
mobility of the roadway system. Input and output measures can be difficult to obtain if the
measure requires data from smart on-board controllers. Plow-down time is the only performance
measure reported that requires (or would benefit from) an on-board system, and only four
agencies reported using this measure, probably because they invested in the on-board control
systems and can now easily report plow-down time. Extensive work has been done to develop
smart snow plows, and equipment that has been tested includes the following:
•
•
•
•
•
•
•
GPS receivers with real-time mapping of truck location
Plow-position reporter
Material dispensed, linked to vehicle location
Pavement temperature sensor
Sensor to detect melting point of road slush
Friction wheel
Heads up display of road center line
These technologies make it possible to report outputs by road segments and link output variables
to vehicle location. The development of smart snow plows is continuing to provide operational
benefits. Successfully measuring friction has obvious implications for performance
measurement, but, recent work by the NCHRP has determined that friction measurement is
experimental. For these reasons, pursuing smart snow plow technology as part of a performance
measure project is not recommended.
Snow and ice control operations also need individual storm information, e.g. precipitation,
intensity, duration, temperature, etc. to be effective. Accordingly, there is a need to quantify the
severity a given storm in order to normalize the efforts expended fighting that storm. Weather
indices are useful for normalizing storm data. Most existing weather indices work on a seasonby-season basis, rather than a storm-by-storm basis. There exists a trade-off between monthly
and seasonal averages: while seasonal data are more normal, fewer observations are available for
analysis. Alternately monthly data offer more observations, but are not as normally distributed.
Maintenance personnel require this information to improve snow fighting capabilities. Seasonal
averages do not provide sufficient data for meaningful analysis for fighting individual storms.
In contrast to input and output measures, outcome measures could measure safety and mobility,
and they affect snow and ice control. Targets for clear pavement or wet pavement by road class
are used by several agencies. Some measures are roughly sensitive to storm characteristics
71
because they identify the maximum allowed accumulation during a storm and maximum time for
removal after snow stops falling. Outcome measures require some form of human observation
that while error can be reduced through training or field guides, human observation always
results in some subjectivity. Additionally, replacing human observation with some form of
technology requires large capital and operational expenses. In the early stages, calibration
problems may actually introduce more variation into the measurements than human observation.
While several technologies are promising, no U.S. agency has been willing to invest in fully
deploying the technology to measure snow and ice control outcomes.
Measures of outcome will help improve safety and mobility. Some form of observation
technology must be deployed to rate outcome measures high in the “measure” and “approach”
criteria. However, widespread deployment is unlikely unless the surface condition data can be
obtained from enhancements to the technology being deployed for other purposes.
To help determine the measures and approaches having potential for use the 11 outcome
measures observed in this study were reduced to three basic categories and two approaches were
identified for each:
1. Measure: Degree of clear pavement
Approach: Manual observation
Approach: Camera-assisted observation
2. Measure: Traffic flow
Approach: Detector-based traffic flow
Approach: Road closure
3. Measure: Crash risk
Approach: Friction (or slipperiness)
Approach: Reported crash data
The Outcome Measure of the Degree of Clear Pavement
Approach: Manual Observation
Measures of the degree of clear pavement by roadway type are common can be used to relate
conditions to safety and mobility. The issue of measuring clear pavement in the winter is similar
to that of measuring pavement distress in the summer as the basis for programming road
improvements. However, snow and ice conditions used to be measured at much shorter intervals
(hourly versus annually).
While it would be desirable to replace human observation with automated technology the fact is
that maintenance supervisors are on duty during storms and can provide condition data at no
additional cost makes it attractive with the use of pictorial guides, a reasonably objective
determination is possible. While potential for improvement in this approach is not envisioned
best practices for manual observation could be compiled as part of a future research effort
72
Approach: Camera-Assisted Observation
No agency reported using freeway-monitoring cameras as an aid to human observation.
However, because most large urban areas are deploying cameras as part of regional traffic
management systems, it is possible that these cameras can be used to observe road conditions on
urban freeways. Maintenance supervisors may not have access to the cameras or they may be on
the road making manual observations anyway, and thus the cameras would not help significantly.
However, the use of cameras appears to be an appropriate measure for snow and ice control.
While the use of cameras is limited, primarily to metropolitan areas, the technology is proven
and can aid agencies in observing road conditions with limited personnel. Furthermore, cameras
are known to being studied to enhance human observation, rather than actually detecting surface
conditions. Detectors are discussed in the following section.
The Outcome Measure of Traffic Flow
Approach: Detector-Based Traffic Flow
This approach considers measures of traffic flow, including speed, volume, and occupancy. The
Ohio DOT and Ada County, Idaho, are the only agencies that reported speed during a storm as a
performance measure. Caltrans and Ada County reported using traffic volume during a storm as
a measure. The Iowa DOT is currently experimenting with the use of ATRs to measure traffic
speed and volume. The following five agencies use time to return to “normal” traffic volume
after a storm as a measure:
• Caltrans
• Detroit Street Maintenance Division
• City of Cedar Rapids, Iowa
• Cuyahoga County, Ohio
• Ada County, Idaho
Despite the lack of widespread use, there appears to be great potential in traffic flow measures.
Traffic speed, volume, and road occupancy are direct measures of mobility. While speeds do not
always drop as much as expected in bad road conditions due to aggressive driving, drivers do
increase the spacing between vehicles, resulting in reduced volume or reduced occupancy.
Traffic flow theory will provide the relationships between occupancy, throughput, and speed.
Currently, cameras and traffic flow detectors are being deployed as components of ITS across
the U.S. Several detectors show promise as tools for measuring road conditions, including
cameras that utilize the visible spectrum and the non-visible spectrum. Other research has
produced cameras that utilize lasers or the infrared spectrum to analyze surface conditions. Such
cameras, some of which are being field tested in the United States, have shown promise in
accurately sensing the presence of differing amounts of water, frost, snow, and ice on a roadway
surface. Side-fire radar has also become quite reliable for measuring speed and occupancy.
73
We recommend pursing detection-based approaches deployed as upgrades to ITS monitoring.
The ITS investment is being made in most urban areas already, and winter maintenance
measurement capability could be added at an incremental cost rather than at full cost.
Additionally, state agencies already collect speed and volume data by deploying ATRs
periodically along the interstate and national highway systems for the routine monitoring of
VMT. Identifying snow and ice event times and capturing the ATR data as the basis for
performance measurement is attractive because it uses an existing investment for a new purpose.
The disadvantages involved in pursuing traffic flow approaches include the following:
•
•
•
Low density of detectors in rural areas
Institutional barriers within DOTs: neither traffic data nor traffic control centers have
a winter maintenance mission; their mission would have to be restructured to deliver
this information to maintenance personnel and, perhaps, to the public
Upgrades to the operations platform that drives the ITS
In general, the lack of technological approaches to measuring snow and ice control performance
indicates that the benefits are not now perceived as worth the investment. By adding winter
condition measurement capability to ITS upgrades and capturing ATR data for this purpose, the
cost of acquiring road condition data can be reduced. The technologies involved are mature
enough to deliver the desired result.
Approach: Road Closure
Road closure is a simple measure of traffic flow, since none is allowed. This is a very useful
measure to record because it is an input for calculating the economic cost of lost mobility. Little
can be done to improve upon it, however. We recommend that road closures be recorded,
possibly as lane-hours of closure. However, the measure is only appropriate when the struggle
against snow and ice has been lost.
The Outcome Measure of Crash Risk
Approach: Friction (Slipperiness)
NCHRP Project 6-14, which resulted in Feasibility of Using Friction Indicators to Improve
Winter Maintenance Operations and Mobility, determined that friction is a feasible quantity to
measure. Thus, we recommend additional work on friction measuring technology, and pursuing
the type of performance measurement that could be based on friction. Ohio is the most
pioneering state on this topic. We recommend documenting more fully Ohio’s activities in
friction-based performance measures.
74
Approach: Reported Crash Data
Crash data collected during winter storm events can provide a basis for measuring performance.
Trends in crashes during snow and ice conditions could be used to measure change in
maintenance performance. Analysis tools are already used in many states that allow users to
search the crash records by weather condition as reported by the officer at the seen of the crash,
and, if correlated with storm severity, provide a fairly robust measure.
One of the early findings of preliminary work by Iowa State University is that crash risk during
the first storm of the season is always far worse than succeeding storms. Apparently, drivers
must relearn winter driving skills each year. (Maze et al. 2005)
As part of the overall safety programs, states continuously improve their collection and analysis
of crash data. The analysis of crash data will likely yield best practices including specific
recommendations for the type of weather data to include in crash reports and the best tools for
analyzing crash data. For example, we know that all crashes that occur during a winter storm are
not reported, but crash databases will include personal injury crashes. We know from prior
analysis of the Iowa crash database that the number of reported crashes skyrockets for winter
storm periods. (Maze et al. 2005).
Storm Severity Index
In addition to the 11 different types of outcome measures reported in the survey, 15 measures of
storm severity were identified in the literature search. We recommend agencies develop an
operational storm severity index that can be applied to normalize any other measure over time.
This would be a great benefit to comparing costs by time period. Our analysis shows that storm
severity models using monthly and seasonal data that the models using monthly data are more
useful for managers fighting storms.
Customer Satisfaction
Customer satisfaction sets the level of performance that the public expects, as a measure of
performance. Because the performance measures measure how close winter road maintenance
comes to meeting public expectations, those outcomes can also be measured by surveying the
public directly. Most agencies use a periodic survey to determine public expectations, and
agencies also track complaints and 511 calls. We recommend defining best practices for
determining customer satisfaction and linking operational performance to those expectations.
5.4. Summary of Approaches
In summary, we recommend the following:
•
Document best practices for manual observation of pavement conditions.
75
•
•
•
•
•
•
Document the use of traffic control center cameras or remote cameras to aid manual
observation inputs to performance measures.
Pursue detector-based approaches that use traffic speed, volume, or occupancy as
means of acquiring data measuring performance.
Document measures that are or can be based on friction.
Document best practices and opportunities for recording and analyzing crash data
during winter storms for use as a performance measure.
Develop a reasonable procedure for incorporating a winter storm severity index to
normalize input, output, and outcome measures.
Determine best practices in the measurement of customer satisfaction and link those
measures to measures of operational performance.
CHAPTER 6. CREATING PERFORMANCE MEASURES TOOLBOX FOR SNOW AND
ICE CONTROL OPERATIONS
6.1. Introduction
Interviews with snow and ice control operations personnel revealed that performance
measurement programs are established for numerous and various reasons. Many reasons focused
on budgetary management, while others were more political, such as legislature mandating
performance. The first step in evaluating effectiveness of performance measures are to determine
why they are in place and to ask what is to be accomplished by instituting a performance
measurement program. This chapter focuses on developing a toolbox to evaluate relevant
performance measures for snow and ice control operations.
6.2. Benefits of Using Performance Measurement
The basic purpose of any measurement system is to provide feedback, relative to the agency’s
goals, that increases the chances of achieving these goals efficiently and effectively.
Measurement thus gains true value when used as the basis for timely decisions.
The accounting firm of Price Waterhouse (Artley and Stroh 2001) has suggested three main
reasons for establishing metrics in an organization that are applicable to snow and ice control
operations.
1. Measurement clarifies and focuses long-term goals and strategic objectives.
Performance measurement involves comparing actual performance against expectations
and setting up targets by which progress toward objectives can be measured.
2. Measurement provides performance information to stakeholders. Performance
measures are the most effective method for communicating about the success of
programs and services. For example, in public education, states and school districts
routinely issue “report cards” highlighting test score outcomes and other key indicators of
76
educational performance. These have become centerpieces of attention among not only
educators, but many other stakeholders. Snow and ice control agencies can also benefit
from “report cards” regarding their performance.
3. Measures encourage delegation rather than “micro-management.” Hierarchical
structures and extensive oversight requirements can hinder organizational effectiveness.
Performance measures free senior executives for more strategic decision making and
collective intervention, while clarifying the responsibilities and authority of managers
down the line.
Organizational metrics are important for these organizations. Working with employees,
management, and affected stakeholders, organizations involved in strategic planning can develop
measures of performance in the production of goods and services and in meeting the
organization’s most important objectives.
There is no single model or process for developing performance objectives and measures, nor is
there a process that will guarantee good results. We have attempted to synthesize lessons learned
from the literature as well as the insights gained from our surveys and work with agencies in
applying performance measurement to the management of snow and ice control operations
issues.
Developing a Performance Measurement Toolbox
One method used to develop performance measurements for snow and ice control is to apply a
toolbox to the problem. A performance measure toolbox brings structure to performance
planning and clarifies the connection between activities, outputs, and results. The toolbox uses
the following steps relative to the objectives specified in an agency’s strategic plan:
Step 1. Confirm Snow and Ice Control Operations Role.
The agency should define the role that snow and ice control operations are intended to play with
respect to strategic objectives and should provide a basis for establishing overall targets and
performance measures. This step will guide the type of goals and objectives to be measured. For
example, if the reason for establishing a performance measurement program is budgetary, then
the measures used will involve ranking investments and allocating resources based mostly on
internal decisions. Externally based reasons may have to do with evaluating the department
against peer comparisons and establishing comparable benchmarks to other peer organizations.
Defining the role that the program is intended to play with respect to strategic objectives
provides a basis for establishing program targets and performance measures. That is, have the
links between the main activities and outputs of the program and the department’s snow and ice
control objectives been established (e.g., activity/output “Y” contributes to, or detracts from,
strategic objective/outcome “X”)? The department’s snow and ice control strategy should
identify the department’s significant snow and ice control aspects as well as its strategic
objectives for addressing these aspects and the measure that will be used to indicate progress.
Table 15 shows the link between the department’s activities and the strategic outputs. Plowing
and anti-icing, for example, directly contribute to mobility.
77
Table 15. Linking program activities and outputs to strategic objectives
Main program
activities or
outputs
Plowing, AntiIcing
Plowing, AntiIcing
Plowing, AntiIcing
Plowing, AntiIcing
Contribution to / detraction from
strategic snow and ice control
objective(s)
Y
Strategic objectives or outcomes
that the program activity or output
contributes to
Mobility
Y
Safety
Y
Productivity
Y
Environmental Quality
For instance, consider the budget-driven model of the Wisconsin DOT, in which the state
reimburses 72 counties perform winter maintenance on state and federal roads (Adams et al.
2003). The three most important current objectives for Wisconsin are as follows:
1. Provide bare/wet pavement in a reasonable amount of time and effort.
2. Improve the coefficient of friction between vehicle tires and the pavement.
3. Provide good winter driving conditions using the most efficient methods possible.
The agency uses a single measure to measure the LOS for snow and ice operations. Periodic field
condition surveys are conducted to measure the traction conditions due to anti-icing chemicals,
sand application, or plowing on the travel lane road surface. These conditions are determined by
observing a mile of road after a winter operation has taken place. Bare pavement is considered
95% of the roadway being free from ice and snow. A roadway is considered sanded if at least
60% of the travel lane has sand on its surface. This is equal to a travel lane with bare tire tracks
with sand on the remainder of the lane (Conger 2005). The data sources for the performance
measurement system include accounting records, visual inspection by law enforcement, periodic
customer surveys, and AVL data from plows.
The measures of these objectives include the time to bare/wet pavement and costs per lane-mile,
while the performance levels have not yet been established. The impetus of the Wisconsin
performance measurement program is thus budget driven. The program developed by the
Wisconsin DOT combines the vehicle data obtained by the on-board systems with weather event
data, labor inputs, and equipment costs, and spatial data. According to Adams et al. (2003), this
system provides information that managers can use to show relationships by vehicle, patrol
section, and storm (e.g., salt application rate, pavement temperature versus weather conditions).
Step 2. Identify the Key Snow and Ice Control Activities and Outputs
This step is to direct winter maintenance managers and staff to identify and focus on the key
program activities. Only those activities that directly relate to the department’s strategic
objective should be measured. Subsequently, only those measures that provide useful
information should be used. Collecting data and information can be time consuming and
expensive so it may be impractical to collect data on every departmental activity.
78
This step is essential to ensure that program managers and staff focus on key issues that
contribute to the achievement of the department’s strategy for snow and ice control. When
establishing the toolbox, it is important to ask whether the key activities and outputs of the snow
and ice control program, in terms of their importance (e.g., high, medium, low) in contributing to
the department’s strategic objectives, have been identified. Table 16 illustrates an example of
this process.
Table 16. Identifying the key program activities and outputs
Program activities
and outputs
Plowing
Output: Road
Condition
Rank, in terms of significance
Strategic objective
Strategic objective
from Table 16:
from Table 16:
safety
mobility
High
High
High
High
Strategic objective
from Table 16:
Productivity
High
High
KDOT, for example, identifies key activities of snow and ice control based on road condition
across road categories. Specifically, KDOT’s objectives were noted as providing a safe travel
way and using resource efficiently. The specific measures that KDOT uses to indicate a safe
travel way vary across three roadway categories, but not across storm type or characteristics.
Data indicating these measures come from reports by field personnel and a computer system
where field personnel record road conditions. The three categories, measures, and performance
levels are as follows:
•
•
•
Category I: two bare/wet wheel paths
Category II: both lanes on two-lane roads with intermittent bare/wet wheel paths
Category III: one wheel path on two-lane roads with intermittent bare/wet wheel
paths
However, while these performance levels measure the road condition, they do not necessarily
indicate a safe travel way. The road condition measure indicates the plowing effectiveness;
however, “safety” is generally measured by the absence of crashes. That measurement is not
indicated among the performance measures.
KDOT currently obtains these observational, road condition data from each district and then
segments the data by district, area, and sub-area for analysis. The data, however, are not
collected on a routinely, timely basis, and much of the analysis is performed well after the storm
event is completed. Thus, there is no immediate feedback provided.
79
Step 3. Identify Program Stakeholders and Issues.
To identify the customers whom the winter maintenance activities and outputs should serve,
influence, or target; the other principal groups affected are; and the ways these groups are
affected. For example, to focus on reasonable access to farms and ranches in rural state might
involve different performance measures than a focus on keeping long-distance roadways clear to
allow reliable and safe movement of freight.
To formulate a set of snow and ice control objectives, it is essential to identify the customers to
be served. Generally, snow and ice control managers have two groups of customers, internal and
external. The internal group is the performance measurement user, generally upper management.
Management uses the performance measurement information as a decision support tool for
budgetary and planning functions. The data gathered from the process help determine program
needs, allocation of funds, and selection of projects.
The external customer is the road user. This is also an important stakeholder. The road user
provides input in the development of the agency’s goals and objectives. In particular, the road
user’s opinions about the strengths and weaknesses of the snow and ice control agency can
influence the agency’s goals and objectives. Many of the snow and ice control agencies we
surveyed indicated that they periodically solicit public opinion to assess their job performance.
By identifying and addressing customer needs performance measurement can help agencies
respond to those needs.
Regarding the main program activities and outputs identified in Table 17, managers must also
determine how the relevant snow and ice control issues affects the stakeholders. Managers must
be aware that the activities and outputs associated with the activities can have both desired and
undesirable effects on the stakeholders. When possible, actions must be taken to mitigate
undesirable effects.
Table 17. Identifying key snow and ice control issues and affected stakeholder groups
Main program
activities and
outputs,
in order of
significance
Example:
plowing
Example:
chemical usage
Snow and ice control issues
Desired
program
effects
Clear roads
Undesirable program
effects
Stakeholder groups
(affected parties)
Positively
Negatively
affected
affected
Inefficient energy use
Road users
Snow and ice Inconsistent application
removal
of chemicals
Road users
Possible
environmental
damage
Possible
environmental
damage
Caltrans, for example, uses a variety of performance indicators for both its internal and external
customers. As input measures, Caltrans uses personnel hours and overtime hours. For input
80
measures, Caltrans uses tons of materials used, lane-miles plowed, amount of equipment
deployed, and cost of winter operations per lane-mile.
Explicit performance levels and standard approaches have not been established for the “time to
bare pavement” measures, although the department does compare predicted weather forecasts
with actual outcomes. Overall, they noted that targets are set annually for meeting performance
objectives in California.
Utilizing performance measurement results in improved communications with staff, improved
decision-making and performance, and improved external communications. If the agency can
identify specific targets and timelines for the measures that have been identified, then a more
comprehensive and effective performance measurement program can be achieved.
Step 4. Identify What the Snow and Ice Control Operations to Accomplish
This step is to illustrate that the results are defined in terms of outcomes that then become the
focus for determining appropriate objectives, milestone targets, and measures (e.g., that
managers receive appropriate feedback).
The organization must establish the results it expects to achieve in terms of outcomes that then
become the focus for determining appropriate objectives, milestone targets, and measures. In this
step, consider whether the desired long-term snow and ice control outcomes for each program
activity or output have been established, document the positive effect(s) that need to be
produced, and consider whether near-term outcomes that can be expected to lead to the longterm outcomes have been established. While this work step often focuses on establishing
objectives to redress identified undesirable outcomes, it is possible that positive effects can also
be further reinforced or improved. Table 18 illustrates an example of this step.
Table 18. Defining results
Desired snow and ice control results (objectives)
Program activities and outputs, Long-term
Near-term
in order of significance
strategic
intermediate
Example: plowing
Maintain mobility
Roadways plowed within
8 hours after storm
Example: chemical distribution
Reduce highway
Treat roadway surfaces
accidents
prior to approaching storm
For example, when asked what performance measures are the most critical to its operations, the
Iowa DOT stated its overall goal is to minimize travel disruptions during winter storms. Current
objectives were listed as safety, returning roads to near-normal driving conditions as soon as
possible, and using the right type and amount of deicing materials at the right place and time; the
department also acknowledged the need to strike a balance between budget, customer service,
and the environment. In terms of specific performance measures, Iowa’s survey response
identified Iowa as one of the states using different performance levels or targets for different
roadway classifications.
81
Step 5. Identify Responses and Performance Requirements.
This step determines how objectives are to be achieved. Performance objectives must be defined
in operational terms to be managed effectively. It is important to consider whether the necessary
performance requirements have been defined to achieve the desired snow and ice control results.
Table 19 illustrates an example of this step.
Table 19. Performance requirements relative to responses and results
Objective(s)
Example: Time to
bare pavement
New or modified
activities, outputs, or other
necessary program response(s)
Plowing, clearing roadway
Performance reqs. relative to
each activity, output, or other
necessary response
Bare lane indicator,
Friction measurement
For example, using friction as a performance measure. The friction coefficient is determined
there are “trigger” values of the friction measurement that would require prompt treatment in
those areas where the friction is less than adequate. Friction measurement devices can detect icy
conditions earlier, provide input for operational decisions in a timely manner, and pinpoint
problem areas with geo-referencing devices.
Establishing Performance Measures
The next four work steps are intended to help snow and ice control operations personnel
establish sound performance measures as well as accountability and resource requirements for
implementation.
Step 6. Identify Potential Performance Measures.
In this step, the agency should develop a list of performance measures that correspond to
performance targets. Performance measurement is required to understand the gap between actual
and expected levels of achievement and the times when corrective action may be warranted. The
results indicated by a performance measure will generally be compared with expectations
specified by a performance target (which might be based on a benchmark best practice, a
technical standard, or some specified progression from the baseline value). Therefore,
performance measures should correspond to performance targets and indicate the extent to which
the organization is achieving these performance expectations. Performance measures are an
important source of feedback for effective management.
The set of measures should address each aspect of the performance framework or toolbox.
Recalling the performance framework outlined above, some performance measures will reflect
how well the program was managed. Such measurements may, for example, focus on time to
reach bare pavement, number of accidents, or costs per lane-mile. Measurement in this area can
involve observation (checking the roadway using traffic cameras) or feedback (e.g., public
comments or survey responses). Finally, some measures will allow a judgment to be made for
82
long-term objectives. These measures involve monitoring long-term efforts such as
environmental impacts (chemical usage, roadway impacts, vegetation impacts, etc.) that can
plausibly be linked back to the program initiatives, outputs, and intermediate effects. These
measures serve as the ultimate barometers of program success. Table 20 provides an example of
establishing performance measures.
Table 20. Establishing potential performance measures
Activities, outputs, or
Objectives
other program responses
Example: Provide winter Plowing, sanding,
storm event response to chemical application
minimize disruption to
normal operations
Example: Provide
efficient snow and ice
control services
Plowing, sanding,
chemical application
Performance
requirements
Time to bare
pavement, time
to bare lane
Costs per lane
mile(fuel,
equipment,
personnel,
chemicals, etc)
Potential
performance
measure(s)
“X” hours after
storm completed
Costs per lane mile
met or exceeded
target
The specific measures used may vary from garage to garage or county to county, based on
geography, weather conditions, road patterns, traffic, etc. The overall organizational goal will be
the same, but specific measures may have to be adjusted to accommodate geographic and
operational conditions.
Step 7. Establish Information Capabilities and a Baseline for Each Measure
In this step, agencies should establish the initial value or baseline of each measure.
Understanding the information currently available to the organization as well as the
organization’s capabilities for gathering and analyzing information is an important first step in
the selection of performance measures. Moreover, establishing baseline measures for each
measure will shed light on the organization’s information capabilities and gaps. Baseline
measures help clarify the implications of objectives in terms of “level of effort” and resource
requirements, and they facilitate assessment of the extent to which progress has been made from
an initial condition. Baseline information provides a further context that helps clarify the
magnitude of performance challenges and achievements. Table 21 provides examples for
establishing baseline measures.
Table 21. Establishing baselines for measures
Potential performance measure
Units
Initial or baseline value
Time to bare pavement
Hours
Hours after storm ends
Return road to near normal conditions
Hours
ADT
83
Step 8. Assess the Adequacy of Performance Measures
Once a list of candidate performance measures has been developed, the next step is to select a set
of performance measures that are suitable for tracking performance toward the specified snow
and ice control objectives. Managers should select requirements that are important to the
organization’s overall goals and performance measures. Screening the set of measures also helps
to ensure that there are no costly measurement redundancies or gaps.
Table 22 is intended to help users assess the candidate performance measures developed in
previous steps. Note that there is a high degree of consensus about the attributes of a good
performance measure. The table summarizes the attributes of a good performance measure and
concludes that good performance measures are meaningful, reliable, and practical.
Table 22. Quality criteria for performance measures
Explanation
Attributes
Meaningful
Understandable
Relevant
Comparable
•
•
•
•
•
•
•
•
clear (clearly and consistently defined)
context (explained)
concrete (measurable)
lack of ambiguity in direction
relates to objectives
significant and useful to the users
attributable to activities
allows comparison over time or with other
organizations, activities or standards
•
•
•
•
•
accurately represents what is being measured
(valid, free from bias)
data required can be replicated (verifiable)
data and analysis are free from error
not susceptible to manipulation
balances (complements) other measures
•
•
feasible financially
feasible to get timely data
Reliable
Practical
For example, Mn/DOT uses bare pavement regain time as a meaningful, reliable, and practical
performance measurement. Based on predefined road classes, a statewide timeframe, in hours,
stipulates when bare pavement should be achieved. These ranges are shown in Table 23
(Keranen 2002).
84
Table 23. Mn/DOT ranges for bare pavement regain time
Road class
Super commuter (>30,000 AADT)
Urban commuter (10,000-30,000)
Rural commuter (2,000-10,000)
Primary (800-2,000)
Secondary (<800 )
Statewide range (hrs.)
SC1 to SC10
U1 to U10
R1 to R10
P1 to P10
S1 to S10
The performance measures must also be screened against criteria for quality considerations.
Table 24 can help agencies assess the quality of a performance measure and determine its overall
value for decision making.
Table 24. A screening tool for quality considerations
Performance measures
that satisfy snow/ice
control criteria
Time to bare pavement
Costs per lane mile
Safety
Mitigate environ. impacts
Meaningful (y/n)
Reliable (y/n)
Und. Rel. Comp.
Y
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Practical (y/n)
Y
Y
Y
Y
Y
Y
Y
Y
Satisfies content
and quality
criteria (y/n)
N
Y
Y
Y
Generally, snow and ice control performance measures should correspond to management
objectives. Therefore many of the performance measure developed in support of overall snow
and ice control strategies will be agency-specific. In the case of “cross-cutting” policy issues
addressed by programs or common operations management issues that departments must
confront, departments may have similar objectives and performance measures. Coordination of
objectives and measures may improve management, reporting, and oversight of the overall
agency’s performance.
Step 9. Establish Accountability and Resources for Implementation
Agencies should establish an accountability system that formalizes the relationship between
results, outputs, activities, and resources. It allows people to see how their work contributes to
the success of the organization and clarifies expectations for performance. Table 25 illustrates an
example of implementing accountability, and Table 26 provides an example of identifying
resource requirements.
85
Table 25. Establishing accountability for implementation
Stage in the accountability process
Program objectives for snow and ice control
Example
measure 1
Objective 1
Example
measure 2
Objective 2
Response 1
Response 2
Measure 1
Measure 2
Responsible party(s) for achieving
objective
Activities, outputs, or other responses necessary
to meet objectives
Responsible party(s) for managing activities or
outputs and meeting the requirements
Performance measure(s)
Responsible party(s) for evaluating measures
Table 26. Identifying resource requirements for implementation
Snow and ice control
objectives
Safety
Level of service
Mobility
Activities, outputs, or other
responses necessary
Plowing, anti-icing
Plowing, anti-icing
Traffic speed
Resource requirements
Human Equipment
Drivers Trucks, plows
Drivers Trucks, plows
Drivers ATRs, WIM,
cameras
Other
In addition to internal accountability measures, several states have an externally oriented, highly
visible performance reporting process. Typical audiences include oversight and policy
commissions, the governor’s office, the legislature, and the public. Agencies typically make their
performance data available and accessible through public reports and the Internet. For example,
WSDOT’s Gray Notebook (http://www.wsdot.wa.gov/accountability/graybookindex.htm) is
presented to the state transportation commission, posted on the Internet, and distributed around
the state to legislators, tribal governments, major media outlets, and transportation interest
groups. The agency’s accountability web site includes a subject index that allows users to see the
results of all published performance measures.
6.3. Conclusions
Achieving reliable and relevant performance data for a snow and ice control performance
measurement program is a large task for any organization. The challenges and problems
associated with performance measurement are multiplied by the unpredictable nature of working
with winter weather.
Complex factors influence the usefulness of performance measures. First, the performance
measures must be perceived as reliable. Straightforward processes are best suited for obtaining
reliable data because complexities can cause variations in reporting. Furthermore, each district or
86
garage should have a clear understanding of what to include and exclude from the performance
measurement program. The program should also involve key people in the creation of
performance target definitions and in the reexamination of existing definitions and measures.
In addition to reliability, relevance is a key ingredient in data use. As discussed, relevance takes
many shapes, and managers and jurisdictions each have their own unique needs. Factors
influencing relevance include managerial control, timeliness, fruitfulness, organizational
capacity, and the organizational philosophy of performance measures. This is not an exhaustive
list, yet it is enough to demonstrate that achieving data use is not effortless.
Agencies may be able to improve their snow and ice control services by measuring the
effectiveness of services they provide. Measuring performance, or the results of services,
provides several benefits. The results can demonstrate value to taxpayers. Knowing the results of
the service allows an agency to tell whether it has accomplished its intended objectives, and, if
necessary, adjust its procedures or practices. Concentrating on results also helps agencies be
more responsive to the needs of their customers and may help agencies communicate more
effectively with taxpayers.
CHAPTER 7. DEVELOPING A FIELD TEST PLAN
The purpose of this chapter is to outline a field test plan for developing a performance
measurement program and examining the tools, best practices, and limitations for snow and ice
control. The field test plan is also designed to help the practitioner understand when to use and
when not to use these tools and practices. In addition, a performance measurement program
encourages progressive changes in snow and ice control practices that will help reduce chemical
usage and mitigate environmental impacts while meeting the safety and mobility needs of
roadway users.
The research revealed the organizational objectives associated with snow and ice control
performance measures that relate to the inputs, outputs, and outcomes of snow and ice control
operations as follows:
•
•
•
•
Accounting for inputs used for snow and ice control
Accounting for outputs accomplished
Operational efficiency
Meeting outcome goals
a. Highway safety
b. Highway mobility
c. Public satisfaction
d. Controlling negative environmental impacts
87
7.1. Measures of Input, Output, and Operational Efficiency
The measures associated with inputs, outputs, and efficiency are based on the accounting system
and shop records: fuel used, person hours, material used, equipment deployed, and plow-down
miles. All of this information originates at the operating unit level and is aggregated up to higher
levels for management review. Since the information originates at the operating unit level, it
makes sense to maintain performance measures at this level.
Measures derived from these data can be used for tactical decision making if the reporting
interval is short. For instance, if data reports are received within about two weeks, shops can
examine the resources used and efficiency in light of recent weather. A road classification
dimension can also be added to measure resources used on different classes of roads. Doing so
requires a truck-based GPS recording system to record the truck’s location and the miles, hours,
fuel, material, etc., by road classification.
Adding a seasonal weather index can also lead to some financial planning parameters, linking
material and labor hours to inches of snow or some other weather severity measure. Because the
number of severe storms cannot be reliably forecasted for upcoming winters, the value of this
index is not clear. However, adding a storm severity index to these measures can be beneficial if
this index were linked to standards for time to bare pavement or similar outcome measures.
Thus, the cost of achieving the standard could be measured and factored into refining the
standards.
An example of measuring inputs, outputs, and efficiency at some level Minnesota’s bare lane
regain time indicator. In Mn/DOT’s case, bare lane regain time is determined to be a relatively
direct measure of snow and ice control effectiveness because management decided that bare lane
regain time was a reasonable direct measure of performance. (Other measures, such as crash
frequency and traffic flow, would be useful as secondary methods of assessing performance.)
Mn/DOT measures bare-lane regain time in hours. These target clearance times for snow and ice
removal provide a relative LOS goal for each road class. The results of the bare lane indicator
are shown on the Minnesota state map for each highway link during a winter storm in Figure 16.
The color on the links indicates whether the objective for that specific segment was achieved
(remember that in Minnesota, goals are dependent on roadway classification).
88
Figure 16. Map of Minnesota districts and regain time
In general, implementing performance measures for snow and ice control requires commitment
from management, labor, and financing. Our discussions with several states revealed that, while
there was a commitment to the idea of implementing performance measures but there was no real
financial data available identifying the benefits and costs of performance measurement. Snow
and ice control departments, however, understand the need for performance measures that can
improve the effectiveness of their operations and better serve the public. A cohesive performance
measurement program, understood by departmental personnel, that supports strategic snow and
ice control objectives enables the agency to organize measures, keep track of results, and take
action to improve results.
7.2. Measures of Highway Safety
The measures of highway safety most often cited are friction measurement and some form of
snow- and ice-related crash reporting. Developing friction measurement as practical as possible
and incorporating those data into its performance measurement program.
89
Another measure that is to be considered is the use of friction measurements to determine LOS.
The objective of using coefficient of friction ratings is to objectively quantify the boundaries for
good-fair-poor ratings and to determine the length of time it would take to recover those ratings
to acceptable values. While equipment does exist to measure the coefficient of friction of a
segment of road, the challenge would be to come up with ratings on a continual basis and to
come up with the rating boundaries. Data issues with coefficient of friction ratings are as
follows:
•
•
•
•
Cost
Frequency of data collection
Benchmarks
Liability implications
To use snow and ice crashes, per VMT as a performance measure the agency must have a crash
reporting system that captures the crash time and location accurately and the reports the crash
information quickly. Second, underreporting in a storm is a problem, because law enforcement
officers are overwhelmed by the number of run-off-road crashes. Present reporting is at best
partial. Finally, crash rate is the most representative measure, but it must be linked to traffic
volume during the storm, not to normal traffic volume. Capturing volume information requires a
dense network of recording stations. States are experimenting with the use of ATRs, weigh-inmotion equipment, and traffic cameras to obtain traffic volume information.
7.3. Measures of Highway Mobility
The measures of mobility include travel time, travel speed, travel time reliability, traffic volume,
and lane occupancy. Travel speed and volume are basic inputs to ITS-based traffic management
systems, and using the information for winter maintenance-related performance measures is a
valuable byproduct. ITS systems are capable of reporting estimated travel times to known points.
If this information is estimated during storms and archived, it can be used to measure travel time
reliability. Speed can be measured in a similar way.
Another method of collecting this information is from ATRs already deployed in most states.
Agencies can collect and archive traffic volume, vehicle type, and vehicle speed from various
locations. Using this type of information, the agency can determine a relationship between travel
time and travel time reliability to weather factors in a snow event such as temperature, snow
amount, and average amount of snow.
Still, there are many unanswered questions about speed reduction in a snow event such as the
impact of traffic volume, type of snow, time of day, driver behavior characteristics. A recent
study by Maze et al. found that during snow days (days when more than one inch of snow fell),
crashes increased and were highly correlated with visibility and wind speed. During low
visibility conditions (visibility of one quarter mile or less) and high wind speeds (winds as high
as 40 miles per hour), crash rate increased to 25 times the normal crash rate. While there are
fewer vehicles on the road during the winter storms, those that remained are much more likely to
be in crash and, as a result, the crash rate skyrockets (Maze et al. 2005). To advance this
90
procedure for implementation in the field, additional extensive research will be required with a
larger sample size. Also, calibration of the regression model will allow this performance measure
to demonstrate more meaningful results to both government agencies and the public.
7.4. Measures of Public Satisfaction
Another method to gauge an agency’s performance is to measure public satisfaction. Measuring
public satisfaction with agency performance will help identify a program’s strengths and
weaknesses. Such research lays the groundwork for improvement.
The collected data must be used to improve the agency’s performance. To be effective, the data
gathered must be used for improvement, not to criticize poor results. Public surveys in several
states and countries are pushing agencies to pursue new and different ways to measure the
quality of winter maintenance services being provided and to identify measurements that
correlate with the customer’s experiences, their perceptions and their expectations in terms of
time to clear roads. Meeting public expectations directly relates to road condition and to tie
maintenance LOS to investment choices in dealing with the department’s funding sources.
The scale of implementation can be wide or narrow, depending on the agency. If a state agency
conducts the public satisfaction research, the scale will be statewide and will focus on the
interstate and state highway system. If a city performs the research, the scale will be more
localized and focused on snow routes and residential streets. The data can be assembled by
garage area or district- or area-wide. By defining response areas geographically, the data may
show where the public ranks one area ranks above another area. The data can then be analyzed as
to the cause of the satisfaction and well performing practices can be passed along to other
garages or districts.
The cost of conducting surveys is a moderate cost, but it can provide great benefits. One of the
benefits of conducting public satisfaction surveys is an increased knowledge of the public’s
expectations, the agency’s performance expectations, ways to measure these performance
expectations, and ways to pay for achieving them. By addressing these issues, it is expected that
agencies can better meet public expectations.
7.5. Measures of Environmental Impacts
Highway maintenance agencies strive to provide safe travel during hazardous winter driving
conditions while keeping traffic delays to a minimum. However, these agencies must also
consider the environmental impacts of snow and ice control operations and the traveling public’s
expectation that high levels of service are to be maintained.
Although there are differing opinions among experts as to the magnitude of damage caused by
the application of salt and other chemicals to roads, it is generally believed that these chemicals
do cause some damage to vegetation, accelerate the corrosion of bridge decks and vehicle
underbodies, and pose a danger to waterways (Transportation Association of Canada 2003).
Another environmental concern is the use of sand and other abrasives, specifically their effects
91
on air quality following storm conditions; dust from airborne particulate matter is generated by
vehicles driving over the applied abrasives.
Regarding environmental issues, snow and ice control agencies are continually challenged to
provide a high LOS and improve safety and mobility in a cost-effective manner while
minimizing corrosion and other adverse effects to the environment. To this end, it is desirable to
use the most recent advancements in the application of anti-icing and de-icing materials, winter
maintenance equipment and vehicle-based sensor technologies, and road weather information, as
well as other decision support systems. Such best practices are expected to improve the
effectiveness and efficiency of winter highway operations, to optimize material usage, and to
reduce associated annual spending and corrosion and environmental impacts (Caltrans Snow and
Ice Control Operations 2005). For instance, the Pacific Northwest Snowfighters Association,
consisting of the transportation agencies in the states of Washington, Oregon, Montana, Idaho,
Colorado, and British Columbia, has strived to “serve the traveling public by evaluating and
establishing specifications for products used in winter maintenance that emphasize safety,
environmental preservation, infrastructure protection, cost-effectiveness and performance”
(http://www.wsdot.wa.gov/partners/pns/).
Caltrans implemented a reduced salt-use policy starting in October 1989, which required
transportation districts to develop specific route-by-route plans (Caltrans 2005). That policy
mandated that “Snow removal and ice control should be performed as necessary in order to
facilitate the movement and safety of public traffic and should be done in accordance with the
best management practices outlined herein with particular emphasis given to environmentally
sensitive areas.” During the first winter, Caltrans reduced salt usage by 62% statewide compared
to the previous winter, helped by improved control of the application frequency of de-icing salt.
Storm tracking with the aid of pavement sensors and miniature weather stations, placed
strategically around the state, give vital information to the counties to maximize their resources
of time and materials. Advances in equipment monitors enable the snowplow truck drivers to be
more effective in treating the roads. Optimizing truck routing can save time and money for
districts through reduction in the “dead-head time” where a truck must return empty to a yard to
refill.
Other Environmental Issues in Snow and Ice Removal Operations
In addition to these strategies thorough training for managers and operators regarding
environmental issues, is also recommended especially in material application. For this reason,
effective training programs must demonstrate the value of new procedures and ensure that
personnel are competent in delivering the new program. This can be a significant shift for longtime winter snow and ice control operators. For instance, the MnDOT developed a performancebased program for reducing application rates, called “Salt Solutions,” that provided operators
with tools and systems for making better application rate decisions. Application rates dropped
when the entire organization actively supported the operators in making better decisions and the
agency took the time to measure and reward improved performance (Broadbent 1999).
92
In general, snow and ice operations are growing in complexity and importance, and the need to
adopt best management practices for environmental issues will only increase. Moreover, more
stringent enforcement of current regulations will probably affect future maintenance programs
significantly. Much of the public attention has centered on mobility and LOS, since commerce
doesn’t stop during snow storms. However, environmental impacts can be incorporated into the
performance measures. Data can be obtained using available RWIS to collect environmental data
and salt impact near environmentally sensitive areas. Lessening environmental impacts will also
require additional training in the application of chemical and other operational practices.
7.6. Conclusions
Although the concepts of performance measurement and performance management have existed
for many years, there is increasing demand that agencies begin to transform their organizations
to institutionalize these practices. This pressure is the result of the convergence of two forces
(ICF 2006):
1. Increased demand for accountability on the part of governing bodies, the media, and the
public in general
2. Mounting commitment of managers and government agencies to focus on results and
work more deliberately to strengthen performance
To meet these pressures, an effective performance measurement and management system links
individual and teamwork behaviors to the organization’s business strategies, goals, and values.
For an organization to achieve its goals, it is essential for each employee to understand
individual roles and responsibility for goal achievement, and there must be continuous dialogue
between leaders and employees to set performance expectations, monitor progress, and evaluate
results. Together, leadership and staff must work to plan, measure and analyze, and manage
performance. These three essential action steps are interlinked and ongoing in an organizational
culture that successfully measures and account for performance.
During the performance-planning phase, the first phase of performance measurement, the
organizational business strategy is defined, including its mission, vision, and objectives, and
specific outcomes required to achieve the overall strategy. Goals and plans for how to measure
achievement must be identified in this step, outputs and measures must be defined and requisite
data collection and analysis processes and procedures must be developed and implemented.
Additionally, and most importantly, employees must come to understand their individual roles
and responsibilities with respect to performance measurements and should be given the
fundamental information, resources, competencies, and motivation to ensure their successful
execution.
In the second phase of the performance measurement process, the measurement and analysis
phase, data that inform areas of success and challenge for the organization are collected and
analyzed. Specific elements and factors that contribute to successes or challenges, along with
new and/or modified information needs and lessons learned, are identified. Once performance
data have been collected and analyzed, they must be effectively managed.
93
The third phase of the process, performance management, is the phase in which solutions to
address identified challenges are developed and implemented, along with mechanisms to ensure
the continuation of program or organizational successes. Additionally, performance
measurement systems and processes may be modified as needed to ensure that information
collected through the performance measurement process is timely, relevant, and sufficient. These
steps then cycle back to performance planning (ICF 2006).
Unfortunately, many snow and ice control agencies have not moved beyond collecting
performance data to utilizing these data to proactively manage the agency. A successful snow
and ice performance program relies on the ability to obtain meaningful data, use these data to
manage the program, and institutionalize these practices so that they become routine. It is
important to promote understanding and support the organizational mission, and demonstrate
commitment to managing for results. Staff must buy into the program and feel empowerment and
continuity. Finally, the results of performance management must be communicated among
relevant stakeholders is crucial to the success of any performance measurement or management
system.
While performance measurement is beginning to become more common, very few snow and ice
control agencies are actively involved in using that data to proactively manage. In other words,
performance measurement has not yet become performance management. Careful planning,
consistent implementation, and thorough communication will help shift the snow and ice control
agency beyond performance data collection to effective performance management.
Suggested Research
In this research, we have attempted to produce a method for snow and ice control agencies that
can easily be used to evaluate appropriate performance measures for snow and ice control
operations. While this research laid out the foundations for such an evaluation, more work needs
to be done in this area, particularly in a real-world application.
At a minimum, more data are needed; specifically, the weather and cost impacts estimated for a
wide range of treatment options should be compared against those experienced in a real-world
execution of the same treatment plans. In this way, the evaluation can be brought into closer
alignment with the reality it seeks to represent.
More work is needed to develop protocols to tie safety with performance metrics. For example,
research needs to be conducted that will explore the relationship between snow and ice control
operations and accident rates, and to find a statistically valid relation between the two. Such a
metric could conceivably be used in future work to construct snow and ice control operations
and schedules which directly seek to minimize predicted accident rates in the road network
without the intermediary step of predicting snow depth or road coverage percentage.
94
REFERENCES
American Association of State Highway and Transportation Officials (AASHTO). 1999. Guide
for Snow and Ice Control. Washington, DC: American Association of State Highway and
Transportation Officials.
Adams, T.M., M. Danijarsa, T. Martinelli, G. Stanuch, and A. Vonderhoe. 2003. Performance
Measures for Winter Operations. Proceedings of the 2003 Annual Meeting of the
Transportation Research Board. CD-ROM.
Al-Qadi, I., A. Loulizi, G.W. Flintsch, D.S. Roosevelt, R. Decker, J.C. Wambold, and W.A.
Nixon. 2002. Feasibility of Using Friction Indicators to Improve Winter Maintenance
Operations and Mobility. NCHRP Web Document 53 (Project 6-14). Washington, DC:
National Cooperative Highway Research Program, Transportation Research Board of the
National Academies.
Anderson, B. 2004. Measuring Winter Maintenance: what’s behind the numbers. Paper presented
at the 2004 Annual Conference of the Transportation Association of Canada, Quebec
City, Quebec.
Artley, W. and S. Stroh. 2001. The Performance-Based Management Handbook. Volume 2:
Establishing an Integrated Performance Measurement System. Washington, D.C.:
Performance-Based Management Special Interest Group (PBM SIG).
http://www.orau.gov/pbm/pbmhandbook/Volume%202.pdf
Audrey, J., J. Li, and B. Mills. 2001. A Winter Index for Benchmarking Winter Road
Maintenance Operations on Ontario Highways. Proceedings of the 2001 Annual Meeting
of the Transportation Research Board. Washington, D.C.: Transportation Research
Board.
Baroga, E.V. 2004. Washington State Department of Transportation’s 2002–2003 Salt Pilot
Project. Paper presented at the Sixth International Symposium on Snow Removal and Ice
Control Technology, Spokane, Washington.
Blackburn, R.R., K.M. Bauer, D.E. Amsler, Sr., S.E. Boselly III, and A.D. McElroy. 2004. Snow
and Ice Control: Guidelines for Materials and Methods. NCHRP Report 526.
Washington, DC: Transportation Research Board, National Cooperative Highway
Research Program.
Boselly III, S.E., J.E. Thornes, C. Ulberg, and D.D. Ernst. 1993. Road Weather Information
Systems, Volume 1: Research Report. Report SHRP-H-350. Washington, D.C.: Strategic
Highway Research Program, National Research Council.
Bourdon, R.H. 2001. Best Practices of Outsourcing Winter Maintenance Services. Richmond,
VA: VMS Inc.
Broadbent, T. 1999. Don’t Overdo It. Roads & Bridges, December 1999.
95
Carmichael, C.G., W.A. Gallus Jr., B.R. Temeyer, and M. Bryden. 2004. A Winter Weather
Index for Estimating Winter Roadway Maintenance Costs in the Midwest. Journal of
Applied Meteorology, 43(11), pp. 1783–1790.
Chen, W., P. Cooper, S. Bath, M. Pinili, and B. Hamilton. 1994. Road Sense Index. British
Columbia, Canada: Insurance Corporation of British Columbia.
Cohen, S.J. 1981. User-Oriented Climatic information for Planning a Snow Removal Budget.
Journal of Applied Meteorology, 20(12), pp. 1420–1427.
Conger, S.M. 2005. Winter highway operations: A synthesis of highway practice. NCHRP
Synthesis 344. Washington, DC: Transportation Research Board, National Cooperative
Highway Research Program.
Decker, R., J.L. Bignell, C.M. Lambertsen, and K.L. Porter. 2001. Measuring Efficiency of
Winter Maintenance Practices. Transportation Research Record, 1741, pp. 167–175.
Finnish Road Administration. 2001. Winter Road Maintenance Policy 2001. Finland: Finnish
Road Administration. http://alk.tiehallinto.fi/thohje/pdf/
winter_road_mainten_policy_2001.pdf.
Giloppe, D., M. Burtwell, S. Bald, and V. Muzet. 2002. Winter Maintenance in Europe: Practice
and Research. Paper presented at the 11th International Road Weather Conference,
Sapporo, Japan.
Harrigan, E.T. 1999. Report on the 1998 Scanning Review of European Winter Service
Technology. Research Results Digest number 238. Washington D.C.: Transportation
Research Board, National Research Council, National Cooperative Highway Research
Program.
Hulme, M. 1982. A New Winter Index and Geographical Variations in Winter Weather. Journal
of Meteorology, 7, pp. 294–300.
ICF Consulting. 2006. Measuring Organizational Performance. Perspectives, Winter 2006.
http://www.icfi.com/Publications/Perspectives-2006/doc_files/organizationalperformance.pdf
Japanese Ministry of Land, Infrastructure, and Transport. 2004. Performance Management of
Road Administration in Japan. Japan: Performance Management Office, Road Bureau,
Ministry of Land, Infrastructure, and Transport. http://www.mlit.go.jp/
road/management-e/e_pdf/0403_1.pdf.
Keranen, P.F. 2002. Optimization of winter maintenance in the Minneapolis-St. Paul
metropolitan area using performance targets. Proceedings of XIth International Winter
Road Congress on new challenges for winter road service. France: World Road
Association (PIARC).
96
Knudsen, F. 1994. A Winter Index Based on Measured and Observed Road Weather Parameters.
Proceedings of the 7th Road Weather Conference. Seefeld, Austria: SIRWEC.
Lee, C. and B. Ran. 2004. A Pilot Study to Measure the Potential of Using Speed Recovery
Duration as a Winter Maintenance Performance Measure. Paper presented at the 2004
Annual Meeting of the Transportation Research Board, Washington, D.C.
McCullouch, B., D. Belter, T. Konieczny, and T. McClellan. 2004. Indiana Winter Severity
Index. Paper presented at the Sixth International Symposium on Snow Removal and Ice
Control Technology, Spokane, Washington.
Maze, T. H, R. Souleyrette, M. Agarwal 2005. Impact of Weather on Urban Freeway Traffic
Flow Characteristics and Facility Capacity, Final Technical Report. Aurora Program
Ames, Iowa; Iowa State University
Minnesota Department of Transportation (Mn/DOT). 2005. Minnesota Statewide Highway
Systems Operation Plan. Saint Paul, MN: Minnesota Department of Transportation.
Niemi, G. 2001. Customer perceptions and expectations of Minnesota’s bare pavement product.
Paper presented at the Ninth Maintenance Management Conference, Washington, D.C.
Nixon, W.A. 1998. Friction as a Tool for Winter Maintenance. Proceedings of the Crossroads
2000 Conference. Ames, Iowa: Iowa State University.
Nixon, W.A. and L. Qiu. 2004. Developing a Storm Severity Index. Paper presented at the 2004
Annual Meeting of the Transportation Research Board, Washington, D.C.
Nixon, W.A. and R. Stowe. 2004. Operational Use of Weather Forecasts in Winter Maintenance:
A Matrix Based Approach. Proceedings of the 12th International Road Weather
Conference. Bingen, Germany: SIRWEC.
Ohio Department of Transportation (Ohio DOT). 2003. 2003 Strategic Initiative Eight: Continue
to Improve Snow and Ice Control.
http://www.dot.state.oh.us/strategicinitiatives/SI2003/03SI8.asp
Olander, J. 2000. Winter road maintenance, the Swedish way. Paper presented at Transportation
Research Board’s Snow and Ice Symposium, Roanoke, Virginia.
Oregon Department of Transportation (Oregon DOT). 2004. 2003 Annual Performance Report.
http://www.oregon.gov/ODOT/CS/PERFORMANCE/docs/2003ODOTPerformanceRepo
rt.pdf.
Osborne, D. and T. Gaebler. 1992. Reinventing Government. Boston: Addison-Wesley.
Permanent International Association of Road Congresses (PIARC). 2006. Japan. Snow and Ice
Databook. 2nd Ed. France: Permanent International Association of Road Congresses.
Perry, A.H., and L.J. Symons (Eds.). 1991. Highway Meteorology. London: Taylor and Francis.
97
Pisano, P. 2004. Heavy accumulation: FHWA, state and local agencies gather ideas from Japan
and Europe that promise improvements in winter road maintenance. Roads & Bridges 42,
p. 3.
Rissel, M.C. and D.G. Scott. 1985. Staffing of Maintenance Crews During Winter Months.
Transportation Research Record, 1019, pp. 12–21.
Smith, J. and T. Adams. 2005. Measures for highway maintenance quality assurance.
Proceedings of the 2005 Mid-Continent Transportation Research Symposium.
http://www.ctre.iastate.edu/pubs/midcon2005/AdamsMaintenance.pdf.
Smithson, L.D. 1998. AASHTO’s Winter Maintenance Program: A Proactive Approach to
International Technology Transfer. Paper presented at the 12th Equipment Management
Workshop, Austin, Texas.
Strong, C. and Y. Shvetsov. 2005. Development of Roadway Weather Severity Index. Paper
presented at the 85th Annual Meeting of the Transportation Research Board,
Washington, D.C.
Thornton, K., G. Hoffman, D. Soltis, H. Balikov, and J. Schaeberle. 2003. PennDOT’s
Environmental Management Program. Transportation Research Circular E-C052:
Maintenance Management 2003, pp. 145–162.
Transportation Association of Canada. 2003. Syntheses of Best Practices for Road Salt
Management: Salt Management Plans. http://www.tac-atc.ca/english/pdf/saltplan.pdf.
Transportation Association of Canada. 2005. Winter Severity Index. Projects in Progress.
http://www.tac-atc.ca/english/projectsandpublications/pro-progress.cfm, as of May 10,
2007 .
TransTech Management, Inc. 2003. Strategic Performance Measures for State Departments of
Transportation: A Handbook for CEOs and Executives. Project No. 20-24. Washington,
DC: American Association of State Highway Transportation Officials, National
Cooperative Highway Research Program, Transportation Research Board, National
Research Council.
Virginia Department of Transportation (VDOT). 2005. Asset Management Best Practices.
Richmond, VA: Virginia Department of Transportation, Asset Management Division.
Washington State Department of Transportation (WSDOT). 2004. Snow and Ice Plan.
http://www.wsdot.wa.gov/maintenance/pdf/Snow_and_Ice_plan.pdf.
Yamamoto, C., K. Kishi, K. Sato, and F. Hara. 2004. Importance of Winter Urban Traffic Issues
and Performance Indicators as Rated by Businesses. Paper presented at the Sixth
International Symposium on Snow Removal and Ice Control Technology, Spokane,
Washington.
98
ABBREVIATIONS
ADT
ATR
AVL
Caltrans
CDOT
DOT
FnRA
GPS
INDOT
ITS
KDOT
LOS
MAP
Mn/DOT
PSIC
RGT
RWIS
SNRA
VDOT
VMT
WSDOT
Average daily traffic
Automated traffic recorder
Automated vehicle locator
California Department of Transportation
Colorado Department of Transportation
Department of Transportation
Finnish National Road Administration
Global positioning system
Indiana Department of Transportation
Intelligent transportation systems
Kansas Department of Transportation
Level(s) of service
Maintenance accountability process
Minnesota Department of Transportation
Pavement snow and ice condition
Road grip tester
Road weather information services
Swedish National Road Administration
Virginia Department of Transportation
Vehicle miles traveled
Washington State Department of Transportation
99
APPENDIX A. WINTER MAINTENANCE OPERATIONS PERSONNEL SURVEY
INSTRUMENT
Survey for Winter Maintenance Operations Personnel
Introduction:
Measuring agency performance is recognized as an important part of a public agency’s mission.
Why measure an agency’s performance? There are four primary reasons to do so.
1. To continuously improve services
2. To strengthen accountability
3. To communicate results of programs and services
4. To provide better information for effective decision making including resource allocation
Iowa State University’s Center for Transportation Research and Education (CTRE) is conducting
a survey for the National Cooperative Highway Research Program (NCHRP) to determine what
types of performance measures are being implemented and how they are working in the area of
winter maintenance operations and snow and ice control.
Please complete this brief survey about your agency’s experiences with implementing
performances measures in snow and ice control operations. It will take approximately 15 minutes
to complete.
Name of Respondent:
Agency:
Telephone:
Email:
A-1
1. For snow and ice control operations, does your agency use: (Please check all that apply)
Internal staff
Private contractors
Contractors with other governmental agencies
2. If contractors, or other governmental agencies, are utilized, do you evaluate contractor, or
agency, performance?
Yes
No
Not applicable
3. Is contractor, or governmental agency, performance linked to payment?
Yes
No
Not applicable
4. Does your agency measure the performance of snow and ice control operations?
Yes
No
Not applicable
4a) If yes, what performance measures do you use? (Please check all that apply):
Time to bare pavement
Time to wet pavement
Time to return to a reasonably, near-normal winter condition
Time to provide one wheel track
Friction or “slipperiness“
Level of service, e.g. traffic flow
Travel speed during storm
Customer satisfaction
Crashes per vehicle miles (or km) traveled
Traffic volume during storm
Time for traffic volume to return to “normal“ after storm
Fuel usage
Lane miles (km) plowed
Personnel hours
Overtime hours
Tons of materials used
Amount of equipment deployed
Miles (km) traveled with plow down
Cost of winter operations per lane-mile (km)
Percent of salt spreaders/controllers calibrated
A-2
Other (please describe):
5. How does your agency decide what items to measure in snow and ice control
operations?
6. What performance measures are most critical to your snow and ice control operations?
7. How frequently do you set targets or objectives for measuring snow and ice control?
Quarterly
Annually
Every 2 years
Other (please describe):
8. Specifically, what are your agency’s three most important current objectives for snow and ice
control operations?
a.
b.
c.
9. What measures do you track regularly on each of these objectives, and what is your
performance level on each? (For example, the measure is time to bare pavement, and the
performance level is 8 hours.)
Measure
A.
B.
C.
Performance Level
10. How do you obtain the data for the performance measurement system? (Please check all that
apply)
Accounting records
Visual inspection by maintenance personnel
Visual inspection by law enforcement
Reports from field personnel
Calls from the public, e.g., via 511
Closed circuit television (CCTV) from freeway management systems
Automated Traffic Recorders (ATR) for travel speed and lane occupancy
Periodic Customer Surveys
Other (please describe):
11. Do the performance measures used by your agency vary with road classification and storm
characteristics?
Yes
No
A-3
11a) If yes, how do they vary? (If yes, please attach examples with returned survey.)
12. Do the performance objectives/targets used by the agency vary with road classification and
storm characteristics?
Yes
No
12a) If yes, how do they vary? (If yes, please attach examples with returned survey.)
13. Do you measure agency performance for managing non-storm events (e.g., blowing snow,
black ice, frost)?
Yes
No
13a) If yes, please describe (If yes, please attach examples with returned survey.)
14. Do you use a storm severity index or similar method for categorizing storm characteristics?
Yes
No
14a) If yes, please describe:
15. Does the agency report the road condition to the public based on the performance
measurement system?
Yes
No
15a) If yes, how do you report the road condition to the public? (Please check all that
apply):
Dynamic message signs
Commercial radio and television
511
Internet website
Other, please describe:
16. Please describe the methods used by your agency to budget, track, and summarize the costs
of snow and ice control and road maintenance?
17. In measuring performance for snow and ice control operations, do you segment the highway
areas for measurement? For example, snowplow routes, mileposts by roadway type, garage
or service areas, or other.
Yes
No
17a) If yes, please describe.
A-4
18. What benefit does your agency obtain from performance measurement? (Please check all that
apply.)
Improved business practices with contractors, e.g., scheduled payments, delivery of
materials, etc.
Improved communications with staff
Improved decision processes relating to snow and ice control, e.g., decisions as to
when to plow, how to plow, how much material to use, etc. are more straightforward.
Improved external communications, such as, with the public, vendors, contractors, etc
Other, please describe:
19. Describe technologies that you have tried for measuring performance and your level of
satisfaction with those technologies. (For example, friction measuring device, global
positioning systems, salinity measurement, video logging, automated traffic recorders (ATR),
others.)
20. Regardless of technology what information do you need that you are not now receiving for
measuring and managing snow and ice control operations?
21. Do you survey the public about the agency’s performance in regards to snow and ice control?
Yes
No
21a) If yes, what do the surveys say about your agency’s performance in regards to snow and
ice control?
22. In your opinion, what three factors most account for your agency’s ability to improve
performance over the past few years?
a.
b.
c.
23. What have been the most significant barriers your agency has encountered in improving
performance?
24. Excluding your own, which agencies around the world stand out as leaders in performance
measurement and/or management, in your opinion?
THANK YOU FOR YOUR TIME AND PARTICIPATION.
You may return the survey, by October 15, 2005 via email to:
[email protected] or you may return the survey via U.S. mail to:
Center for Transportation Research and Education
Iowa State University
2901 S. Loop Dr., Suite 3100
A-5
Ames, IA 50010-8634 U.S.A
Tel: 515-294-8103
or FAX your completed survey to: 515-294-0467
If you have any questions regarding this survey, please contact Steve Andrle at CTRE at 515294-8103 or email to: [email protected]
A-6
APPENDIX B. RESPONSES TO SURVEY
This appendix summarizes the responses received for the survey presented in Appendix A.
Responses are organized by question.
1. For snow and ice control operations, does your agency use: (Please check all that apply)
Internal staff
1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26
Private contractors 2,5,7,8,16,17,18,19,26
Contractors with other governmental agencies 5,9,13,17,24,25
2. If contractors, or other governmental agencies, are utilized, do you evaluate contractor, or
agency, performance?
Yes 2,7,9,13,16,17,18,20,24,25,26
No 5,8,19,
Not applicable 1, 3,4,6,10,11,12,14,15,21,22,23
3. Is contractor, or governmental agency, performance linked to payment?
Yes
2,7,9,13,16,17
No
5,8,18,19,20,24,25,26
Not applicable 1,3,4,6,10,11,12,14,15,21,22,23
4. Does your agency measure the performance of snow and ice control operations?
Yes
1,2,3,4,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,26
No
25
Not applicable 5
4a) If yes, what performance measures do you use? (Please check all that apply):
Time to bare pavement 1,9,13,17,18,21
Time to wet pavement 1,3,20
Time to return to a reasonably, near-normal winter condition 4,9,11,12,16,17,19,22,24,26
Time to provide one wheel track 24
Friction or “slipperiness“ 1,4,16,22
Level of service, e.g. traffic flow 1,2,4,9,12,15,19,24
Travel speed during storm 1,4
Customer satisfaction 1,2,4,7,9,10,11,12,13,14,16,17,20,22,26
Crashes per vehicle miles (or km) traveled 4,
Traffic volume during storm 1
Time for traffic volume to return to “normal“ after storm 1,2,22
Fuel usage 11,14,19,20,22
Lane miles (km) plowed 1,2,4,10,11,13,14,16,17,19,20,24,26
B-1
Personnel hours 1,2,4,8,9,11,13,14,15,16,17,19,20,22,24,26
Overtime hours 1,2,4,8,9,11,13,14,15,16,17,19,20,22,24,26
Tons of materials used 1,2,4,8,9,10,11,12,13,15,16,17,18,19,20,22,24,26
Amount of equipment deployed 1,2,4,8,9,11,12,13,15,16,17,19,22
Miles (km) traveled with plow down 9,14,17,26
Cost of winter operations per lane-mile (km) 2,4,8,9,10,11,14,17,18,19,20,24,26
Percent of salt spreaders/controllers calibrated 1,9,11,13,17,19,26
Other (please describe): 3,4,8,18,24
3-On lower volume roads (<1000 AADT) time from the end of the storm to plow roads open
to two way traffic and treat all hills, curves, intersections and other critical areas
4-Formal process for annual Quality Assurance Reports (QAR) on a statewide basis
6-Level of service, pavement conditions for the category of route
5-The State of SD does not formally track performance measures for winter maintenance.
Several of the above are informally looked at. Time to return to reasonably, near-normal
winter condition; traffic flow; and customer satisfaction.
8-Time since route reported treated
18-Contract trucks reporting for winter storms in a timely manner as described in SHA’
Hired Equipment for Snow Removal Services Contract.
24-Material, labor, and equipment hours and costs are tracked
5. How does your agency decide what items to measure in snow and ice control
operations?
1. Resources, safety
2 Based on customer indicators and fiscal barriers
3- N/A
4- Existing guidelines and evolving technology (we are testing friction as a method of
measurement)
5-n/a
6-it was determined to use data that was already being captured (road condition information)
7-Internal decision
8-We measure traditional inputs and outputs from typical maintenance management systems.
We’ve considered others but budget restraints have kept up from moving to those
areas.(AVL/GPS, return to baseline travel speeds, etc.)
9- Determined with input from statewide snow and ice committee and locally at the regional
level through written operational plans.
10-Budget-planned vs. actual cost for Snow and Ice. The standard cost per mile v. actual cost
per mile.
11-Customer satisfaction, snow policy, time of response, intensity, length, and geographic
location of storm, level of service and classification of roads.
12-Snow policy manual
13-Most are recorded automatically and reviewed. Many of the above are being looked at for
future applications.
14-Planning tools needed
15-Past practice
16-Normally, we use the above criteria for all storms.
17-Based on historical stats and customer expectation for a winter city. Maintenance
B-2
Management system outcomes and also derived from media questions as well. If you
measure it you use it to develop K.P.I.s!
18-Performance measures on bare pavement, material usage and cost of operations are
detailed in SHA’s business plan which is developed by senior SHA managers. Performance
measures for hired equipment is determined by Statewide Maintenance Quality Council
which is composed of district and statewide maintenance managers.
19-Budget/political driven, environmental compliance, new technology
20-Selected as part of overall department performance measures. Also selected to help keep
operators informed on their effectiveness and usage.
21-As stated in policy and procedure manual
22-Historical & Industry-wide standards
23-measure materials, personnel and amount of equipment used
24-Several item collected from our Winter Supervisor daily reports and weather information
(RWIS/AWOS) are used to measure certain levels of winter operational performance. We
continue to explore a number of other information sources (Speed, friction, crashes, weather
indices, etc.) to determine if they are good indicators of performance. In the last two years
speed data from the existing ATRs have been evaluated for measuring performance. The
Department has also used customer survey information in the past as well as a survey of
Highway Patrol Troopers.
25-Final decision by Dept. Commissioner
26-Capability to measure and if the information is useful or not.
6. What performance measures are most critical to your snow and ice control operations?
1. safety
2. Customer satisfaction and cost to provide service
3. n/a
4. Customer feedback, pavement conditions, traffic flow, resource deployment
5. n/a
6. road condition
7. travel speed during event and time to complete clean up after event
8. Time since route was reported treated (are all routes serviced and how long since last
treatment.
9. Level of service, customer satisfaction, time to normal conditions, and after storm clean
up.
10. Lane miles cleared and actual v. planned (standard) unit cost.
11. Customer satisfaction and safety
12. time to return to normal winter driving conditions
13. Pavement conditions during and following each event and materials used.
14. Dollar related items and customer satisfaction
15. Condition of the streets and how traffic is moving
16. Time to remove snow and amount of slick areas remaining.
17. Bare pavement and/or customer satisfaction.
18. Time to reach bare pavement
19. Time to return to a reasonable near-normal winter condition
20. Salt per lane-mile and miles plowed.
21. Time to bare pavement
22. Public Safety
23. Public Safety and customer satisfaction
B-3
24. Any measure that shows a direct impact to travelers such as speed, traffic volume,
crashes, etc. Our goal is to minimize travel disruptions during winter storms. If we can
maintain 70 mph speeds, traffic volume is not reduced and cars are not in the ditch or
having crashes, we would call that successful performance. We also need to strike a
balance between our budget, level of customer service and the environment.
25. Time to return to reasonable winter pavement condition.
26. Amount of time to plow all streets
7. How frequently do you set targets or objectives for measuring snow and ice control?
Quarterly 24
Annually 1,2,4,7,8,9,10,12,13,14,16,17,18,19,20,21,22,23,24,25,26
Every 2 years
Other (please describe): 11,15
3-As deemed necessary
4-Ongoing-at various levels
5-n/a
611-semi-annually
15-Objective is monitored regularly
8. Specifically, what are your agency’s three most important current objectives for snow and ice
control operations?
a.
1- safety
2- to maintain passable roadways for emergency vehicles
3- Have all major highways restored to wet or dry condition
4- To provide bare pavement as soon as possible and practical
5-n/a
6- provide safe travelway
7-reduce highway fatalities
8-pre-treat city routes to prevent bonding
9-Providing reasonably safe and clear highways during and after the storm for the
traveling public.
10-Provide for public and employee safety
11-Response time
12-Safe driving conditions
13-Anti-icing
14-Maintain all roads in passable condition
15-Keep traffic moving
16-Plow all County Highways
17-Bare Pavement Policy
18-Provide safety mobility for motorists which allows our customers to carry on their day
to day activities and our business community to remain operational during and after
winter storms.
19-Level of service: time to normal winter driving
20-Safety
21-Obtain bare pavement in minimum amount of time
B-4
22-Safe Roads
23-public safety
24-Safety
25-Keep Roads Clear
26-Time to plow all streets
b. 1- traffic backup
2- To return traffic flow to normal as quickly as possible
3- Have all minor highways greater than 1000 AADT restored to a wet or dry condition
4-Continue to be effective and efficient
5-n/a
6-efficient use of resources
7-customer satisfaction
8-provide continuous service from pre-storm through post storm to minimize the negative
on travel time, safety, etc
9- Maintaining an adequate level of service throughout and soon thereafter a storm.
10-Clear priority routes
11-Customer satisfaction
12-Cost effective snow and ice control
13-24 hour coverage
14-Prioritize efforts by road class
15-Use cost effective procedures
16-Plow all residential streets
17-Ready for Rush Hour (M–F)
18-Protect the environment
19-Cost and budget stability
20-Effective
21-n/a
22-Time utilization
23-driving conditions
24-Return roads to near-normal driving conditions as soon as possible
25-Traffic flow
26-Keep all streets safe for vehicular traffic
c. 1- Time
2- To operate as efficiently as possible
3- Have all minor highways with less than or equal to 1000 AADT open to two way
traffic and treated with salt and/or abrasives on all hills, curves, intersections, and other
critical areas
4-Maintain excellence while considering environment and economic factors
5-n/a
6-n/a
7-keep traffic moving safely
8-provide clear pavement on city routes as quickly as possible
9-Clear roadways within 2 hours after the storm. Under “modified“ level of service, this
can be up to 3 hours.
10-Cost effective snow removal
11-Safety of the traveling public
B-5
12-Customer satisfaction
13-Reduce salt usage
14-Shift resources for best effect
15-Do not harm environment
16-Spread all roads with ice control materials
17-Sidewalk clearing to meet Bylaws (48Hrs)
18-Provide services at the least cost to taxpayers
19-Salt tracking, better usage practices, new materials (pre-wetting)
20-Efficiency
21-n/a
22-Cost effectiveness of supplies and equipment
23- school zones
24-Use only right amount and type of de-icing materials at the right time in the right
locations to get the job done
25-Accident free
26-Meet customers expectations
9. What measures do you track regularly on each of these objectives, and what is your
performance level on each? (For example, the measure is time to bare pavement, and the
performance level is 8 hours.)
Measure
Performance Level
A. 1. N/A
2. Passability for emergency vehicles
2 continuous
3 Time to wet or dry condition
3 As soon as possible after end of storm
4-Performance levels not used
5-n/a
6-safe travelway for category I routes
6-bare/wet wheel paths
7-Time to clean up in urban areas after storm event stops 7-18 hours
8. Pre-treatment
8-Starting 3 hours before beginning of storm
9-Monitor police and public observational/calls
9-minimal complaints/calls
10-Number of lane-miles cleared
10-n/a
11-Response time
11-1 hour
12-Safe driving conditions
12-Equipment dispatched to achieve goal
13-n/a
14-Maintain all roads in passable conditions
14-n/a
15-Labor hours
15-n/a
16-plow county highways 2 passes
16-12 hours after snow stops
17-Areterials
17-48 Hours
18-Bare Pavement
18-Reaching wet or dry pavement within 8
hrs of the ending of frozen precipitation
19-Salt
19-Annual Usage
20-Safety
20-none set
21-Time to bare pavement
21-Minimum time
22-Related Safety forces
22-Feel roads are safe
23- Percentage reports on roadway conditions
23-2 – 3 hours to bare pavement
B-6
24-On high volume roadways, return roads to
reasonable, near-normal conditions within 24
hours
25-Nothing specific
26-Time to plow all streets
24- 95%
26-12 hours
B.
2 Time to normal traffic flow
3 Time to wet or dry condition
4-n/a
5-n/a
6-safe travelway for category II routes
2- 24 hours
3- As soon as possible after end of storm
6-both lanes on two-lane roads with
intermittent bare/wet wheel paths
8-Continuous service
8-person hours and material use
9-Mgt/staff perform patrols during storms to monitor conditions
9- prevent hardpack
10-Number of citizen action requests
10-n/a
11-complaint
11-number of calls
12-Cost effective snow and ice control
12-Entire city plowed 14-16 hours following
conclusion of snow event
13-n/a
14-Priortize efforts by road class
14-n/a
15-Sand & Control products applied
15-n/a
16-Plow residential streets 1 pass
16-8 hours after snow stops
17-Collectors (Bus Routes)
17-48 Hours
18-Protect the environment by using least amount 18-Unknown-storms never
of materials for each storm
same
19-Costs
19-Budget Levels
20-Effective
20-8 hours to wet pavement
21-n/a
21-n/a
22-n/a
`
22-n/a
23- City plow and salted one cycle
23- Every 4 hours
24-On low volume roads, provide bare wheeltrack
within 24 hours
24-85%
25-Nothing specific
26-Visual observations surveys
26- A thru F
C.
2- Operational efficiency
3- Time to open to two way traffic
4-n/a
5n/a
6- safe travelway for category III routes
2- Regional standards
3- As soon as possible after end of storm
6-one wheel path on two lane roads with
intermittent bare/wet wheel paths
8-no specific criteria on time Depends on
each storm
9-2 hrs
10-n/a
8-clear pavement
9-Clear Roadways
10-Unit costs
B-7
11-materials
11-amounts
12-Customer satisfaction
12-positive phone calls compare to negative
13-n/a
14-Shift resources for best effect
14-n/a
15-How traffic is moving
15-n/a
16- Plow residential streets 2nd pass and
16-24 hours after snow stops
spread ice control materials
17-Sidewalk
17-48 Hours after last snowfall
18-Provide services at the least cost to taxpayers
18-Moving target, storm
by using least amount of materials, equipment,
never same
and labor for each individual storm
19-Levels of Service
19-Hours
20-Efficiency
20-200 pounds per lane-mile per storm
average for season
21-n/a
21-n/a
22-n/a
22-n/a
23-Visual inspection of plowing operations
23-n/a
24-On low volume roads, return roads to
24- 95%
reasonable near-normal conditions within 3 days
25-Nothing specific
26-Customer surveys
26- 80%+ Happy with service levels
10. How do you obtain the data for the performance measurement system? (Please check all that
apply)
Accounting records 2,4,9,10,11,12,13,14,15,18,19,20,24,26
Visual inspection by maintenance personnel
1,4,7,9,10,11,13,14,15,16,17,18,19,20,21,22,25,26
Visual inspection by law enforcement 1, 2,4,14,17,19,22,25,26
Reports from field personnel 1,2,3,4,6,9,10,11,12,13,14,15,16,17,18,19,20,21,22,24,25,26
Calls from the public, e.g., via 511– 4, 10,11,12,13,15,16,17,19,26
Closed circuit television (CCTV) from freeway management systems 1, 2,9,15,26
Automated Traffic Recorders (ATR) for travel speed and lane occupancy 9,17
Periodic Customer Surveys 2, 11,14,20,26
Other (please describe): 18,23,26
4-testing friction measurements
5-n/a
6-the computer system where our field personnel record road conditions is used for the
snow and ice performance measures
8-Work Management System (may be same as accounting records.)
18-SHA’s Emergency Operations Reporting System (EORS) and Scan Web (RWIS) data
23-GPS
26-AVL information
11. Do the performance measures used by your agency vary with road classification and storm
characteristics?
Yes 1,3,4,6,7,14,16,17,18
B-8
No 2,8,9,10,11,12,13,15,18,19,20,21,22,23,24,25,26
5-n/a
11a) If yes, how do they vary? (If yes, please attach examples with returned survey.)
4-Areas are tracked-not necessarily rated
6-see question #9
7-right now, performance measures are only in place in urban areas
14-Higher classifications get priority and first attention
16-Larger snowfall amounts and/or ice storms require longer times for removal
operations. The time of day can also affect snow removal operations, i.e., during rush
hour.
17-Time and level of service for sanding, plowing, and snow removal
18-Bare pavement performance measure is applicable only to interstate and primary
highways, not secondary roads, however, the other two measures (environmental
stewardship and cost-effective operations) are the same on all roads.
12. Do the performance objectives/targets used by the agency vary with road classification and
storm characteristics?
Yes 4,7,9,13,15,17,18,19,23,24
No 1,2,3,6,8,10,11,12,14,15,18,20,21,22,25,26
5-n/a
12a) If yes, how do they vary? (If yes, please attach examples with returned survey.)
4-Same as #11
7-higher class roads receive more dedicated efforts
9-Interstates and major arterials are higher priority over secondary highways.
13-Higher AADT roads receive extra help in major storms.
15-We don not sand/plow non arterial street except for emergency routes to hospitals,
etc.
17-Time and level of service for sanding, plowing, and snow removal.
18-Bare pavement performance level of 8 hours is applicable only to interstate and
primary highways, not secondary roads, however, the other two levels (environmental
stewardship and cost-effective operations) are the same on all roads.
19-Time requirements for treatment completed on different levels of roads.
23-Major storms that require snow removal and towing of parked cars
24-High volume roads have higher performance measures than lower volume roads.
Interstates and other higher volume roads are required to return to near-normal within 24
hours while lower volume roads must be returned to near-normal within 3 days.
13. Do you measure agency performance for managing non-storm events (e.g., blowing snow,
black ice, frost)?
Yes– 4,17,19,23,24,25
No 1,2,3,6,7,8,9,10,11,12,13,14,15,16,18,20,21,22,26
5-n/a
13a) If yes, please describe (If yes, please attach examples with returned survey.)
B-9
4-Perform annual QAR’s
17-Blowing snow and freezing rain are considered storm events
19-Miscellaneous winter maintenance: road inspections, snow fencing, sweeping winter
sand, etc.
23-We measure all public work functions relating to snow operations.
24- Our definition of precipitation start and end times is the time when precipitation
accumulates on the roadway surface. This would allow blowing snow events to be
included in precipitation days- In Iowa snowfall may stop but it will be followed by
several hours of blowing snow conditions that accumulate on the roadway. Therefore we
defined start and end times to include any events that precipitation accumlates on the
roadway surface. Bridge frost events are handled separately and are primarily a measure
of the performance of our forecast service. We ask our garages to report whether or not
frost was forecast. If forecast they are asked, was frost found, was frost found on
adjacent bridges (county/city bridges), was frost not observed (used on weekedns when
no one is around to look at bridges.
25-Same as storm events
14. Do you use a storm severity index or similar method for categorizing storm characteristics?
Yes – 4,14,15,17,24
No 1,2,3,6,7,8,9,10,11,12,13,16,18,19,20,21,22,23,26
5-n/a
14a) If yes, please describe:
2- We have not in the past, We plan to implement an index
4- Typically “record“ indicators-such as record snow fall amounts
6- We use the Winter Index from SHRP H-350 and data received daily from NWS
15-Our planned response is based on how much show is forecast, what time of day, and
day of week.
17-Not storm specific but overall winter is reported to Environment Canada annually.
24- We are just starting to use a weather severity index to evaluate winter performance
and are looking at a number of different weather data to measure severity. We have used
the SHRP index, one developed by Mike Adams in Wisconsin and are currently looking
at including data from our daily reports to provide more detailed weather information
than is currently available from other weather resources.
15. Does the agency report the road condition to the public based on the performance
measurement system?
Yes 2,4,6,9,10,13,18,19
No 1,6,7,8,11,12,14,15,16,17,20,21,22,24,25,26
3 N/A
5-n/a
15a) If yes, how do you report the road condition to the public? (Please check all that apply):
Dynamic message signs 2,9
B-10
Commercial radio and television 2,9,13,14,19
511- 6,14,19
Internet website 2,4,6,9,10,13,14,18,19
Other, please describe:
13-Internet site begins this season showing where equipment has been. Radio stations call us,
we do not contact them.
14-Not really a system
16. Please describe the methods used by your agency to budget, track, and summarize the costs
of snow and ice control and road maintenance?
1. Annual budget
2 Prior to 2003, costs nor performance were reviewed. We have implemented performance
reviews to evaluate supply use, communication (internal/external), coordination between
adjacent agencies (i.e. MDOT, Wayne County, and total costs.
3 Budget: Recent five year average cost. Track: Financial Management System. Summarize:
End of fiscal year Expenditure Report.
4- Internal cost accounting system
5-SD tries to budget for what is considered a “normal“ winter. Through the budgeting
process, estimates are made as to the number of man-hours, equipment hours and materials to
be used in the upcoming season. These are tracked through the payroll and inventory
systems.
6-We use a Highway Maintenance Management System
7-We identify and track all snow and ice control activities
8-Labor, equipment and materials are recorded in our work management systems (Hansen
Infrastructure Management System). In addition, work orders are opened and closed as
drivers begin and complete treatment on specific snow routes or if they are just patrolling or
standing by. Reports are generated either direction from the IMS or via specialized report
writing software. Because of variability of winter operations, budget are set on historical use
rather than on influences of a specific year.
9-Data entry into computer program that tracks maintenance activities, personnel hours,
materials, etc.
10-We use a Maintenance Management System based on inventory and history to develop
plans and track actuals.
11-Accounting department and supervisory daily time and after action meetings
12-Previous year totals, current year totals.
13-Materials are tracked for each event and totaled for an annual cost. On board computers
collect data and transfers then information wirelessly to a PC where the data are put into
report form. Data and GPS are plotted to a map and can be reviewed by staff. Time for each
event and total number of personnel needed are recorded but OT is kept in accounting.
14-Maintenance Management system
15-Our budget is developed with an adequate amount to deal with the small winter event. If
the snowfall is frequent or heavy we will run over our budget and ask for emergency funds if
needed. We compile costs for the response on a daily basis using info from the daily truck
sheets from the crews.
16-Work order/cost accounting
17-SAP
18- SHA's snow and ice control annual winter budget is based on an average of previous
winter expenditures. After the budget is determined, costs are tracked by SHA's Maintenance
B-11
Operations Support Team using SHA's EORS and the Financial Management Information
System (FMIS). The team produces reports for senior SHA management throughout the
winter season detailing expenditures for materials, personnel and equipment. Reports on the
performance measures are prepard by SHA's Quality Assurance Team and distributed to
senior managers as well as frontline maintenance managers. At season's end, the
Maintenance Operations Support Team prepares a report on total season expenditures which
is used as justification for a budget amendment if the agency overruns the original budget.
The Quality Assurance Team prepares a report that details the agency's performance
throughout the winter at the shop, district, and statewide levels. The report also details each
storm and reviews the performance of the weather service provider.
19-Daily field reporting of equipment usage, employee time, material quantities used km
treated into a computer database (CODES), Common Data Entry System.
20-All material and equipment usage is recorded daily on timesheets and entered into
accounting system
21-Cost accounting
22-Payroll and supply records
23-Excel sheet for every storm complete costs
24- Each maintenance garage completes a winter daily report on their operations. The report
provides a detailed account of the materials, equipment, crews and weather for the day. Cost
and hours of operations data is also analyzed to measure operational efficiencies.
25-Set budget by category
26-Data review
17. In measuring performance for snow and ice control operations, do you segment the highway
areas for measurement? For example, snowplow routes, mileposts by roadway type, garage
or service areas, or other.
Yes 3,6,8,9,11,12,13,15,16,17,18,19,23,24,25,26
No 1,2,4,7,10,14,17,21,22,
5-n/a
17a) If yes, please describe.
3 - By major and minor highways and by maintenance building, superintendent area,
district and statewide results.
6- By our District, Areas, Subareas, and highway routes
8- Indianapolis operates out of 3 main garages. Costs for operations are tracked
individually and related back to the number of lanes miles of responsibility in each area.
We do not segment to a finer detail than that.
9- Plow beats, and highway corridors
11- snow plow routes and segmented service areas
12- 95 snowplow routes
13-By route which averages 28 miles each. We maintain 550 lane-miles.
15-We have developed routes that can be handled by a truck using one load of sand for
each route.
16-Major snow routes are numbered and residential areas separated.
17-We inventory and route everything.
18-Performance is measured at the maintenance shop, district, and statewide levels.
Performance at this time is not measured at the snow route level.
19-Level Classifications based on AADT and DHT Classification system.
B-12
23-We have 15 snow districts and also fleet mgt services.
24-Performance measured by service level of the roadway segment and the garage area
responsibility.
25-District and road
26Lane miles
18. What benefit does your agency obtain from performance measurement? (Please check all that
apply.)
Improved business practices with contractors, e.g., scheduled payments, delivery of
materials, etc.
2,11,13,14,17,18,23
Improved communications with staff
2,3,4,8,9,10,11,12,13,17,20,22,26
Improved decision processes relating to snow and ice control, e.g., decisions as to
when to plow, how to plow, how much material to use, etc. are more straightforward.
1,2,4,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,24,25,26
Improved external communications, such as, with the public, vendors, contractors, etc
1,2,7,8,9,10,11,12,13,16,17,19,26
Other, please describe:
20
3-The benefit of knowing how well we deliver services and products to our customers.
4-Uniformity of service across jurisdictional boundaries
5-n/a
6-Our current performance measures are very time consuming to generate on a timely
basis and are currently not being used very widely.
14-Effectiveness of products applied per situation
20-operators do a better job knowing expectations are more than just “plow the road”
19. Describe technologies that you have tried for measuring performance and your level of
satisfaction with those technologies. (For example, friction measuring device, global
positioning systems, salinity measurement, video logging, automated traffic recorders (ATR),
others.)
1 field personnel
2 N/A
4-Currently testing friction measurements-satisfaction undetermined
5-n/a
6-n/a
7-n/a
8-We’re hoping to utilize some of these in the future, most specifically AVL/GPS and
ability to track vehicle speed. We haven’t used any of these to date.
9-Currently piloting AVL, RWIS (unfortunately RWIS has not been reliable), increased
use of pre-treating with liquids.
10-n/a
11-Weather radar but will be setting up ATR
12-n/a
B-13
13-GPS, automated recorders in the trucks.
14-n/a
15-n/a
16-n/a
17-n/a
18- Currently, SHA uses visual inspection by field personnel as a primary tool for mesuring
bare pavement performance. Office staff uses RWIS data to supplement and at times validate
bare pavement performance. Our performacne measure would be more valid if it came from a
more objective source other than the folks who are actually peforming the work.
19-GPS, material tracking through hydraulic systems in new tandems, (km treated, material
type, speed of truck, etc.)
20-AVL in 3 trucks at another County
21-None
22
23-We are in the process of implementing gps on snow contractors for the 2005-2006 season
24- We have tried friction wheels but found them to be too costly. We are currently using
speed information from ATR's but hope to also install additional speed sensors at RWIS sites
in the future to combine weather and traffic information to use as a measurement. We also
have worked with Iowa State University to use artificial intelligence to tie budget with
weather severity for use as a measurement of performance. We also are working on crash
data to determine if it can be used to measure performance.
25-None
26- AVL/Time and Labor Tracking Software
25. Regardless of technology what information do you need that you are not now receiving for
measuring and managing snow and ice control operations?
1 N/A
2. Radio reports from cities in the storm’s path ahead of us (180 arc) on consistency,
duration, temp., wind
3-N/A
4-Improved performance measure indicators/processes, improved communications
5-A better handle on the amounts of chemicals used and the results of those applications.
Efficiency needs to be realized in the use of the chemicals-more is not always better.
6-Consistent data source of snow fall amounts. It is sporadic data available from NWS.
7-automated cycle time and amount of materials being expended
8-Specific equipment functions such as spreader rates, plow-up/down and plow speed.
9-Reliable RWIS information
10-GPS tracking
11-ATR
12-GPS
13-We have many areas in which to improve. We need a benchmarking study.
14-We want better RWIS coverage
15-We have tools we need
16-AVL
17-RWIS road temperature measuring and incident detection with live cameras
18- Currently, we do not have a means for capturing an average pavement temperature
during winter storms at the shop, district and statewide levels. We can determine the pounds
of salt used per lane-mile per inch of snow from RWIS data but pavement temperature
B-14
should be factored into the equation. It can be done manually but it would be an extremely
time consuming operation. SHA RWIS programmers are currently working on this issue.
19-storm conditions and severity, real time road condition data (pavement
temp/precipitation.)
20-Actual time plowing better time to wet pavement data
21-None
22-n/a
23-The throwouts of snow back into the street when residents are shoveling there cars out is
a major problem we have a problem managing this due to the volume of parked cars and the
congestion of the streets.
24- Speed of traffic and impact (incovenience, delays, etc.) to roadway users. The airline
industry understands the cost of shutting down and airport for a certain period and the impact
it has on the system. Based on that information, airports started using sophisticated
equipment and deicing materials to minimize down time for runways. Unfortunately there
doesn't seem to be similar information available about the impact of snow and ice to the
travelling public that could be used to lobby for more funding or resources for use on
roadways.
25-Pavement temperature
26-Friction data
26. Do you survey the public about the agency’s performance in regards to snow and ice control?
Yes 2,5,6,7,11,12,14,17,18,20,23,24,26
No 1,3,4,8,9,10,13,15,16,19,21,22,25
21a) If yes, what do the surveys say about your agency’s performance in regards to snow and
ice control?
5-It is good, but it can always be better. The highest priority amongst the public surveyed.
6-The survey was conducted several years ago and the public was very pleased.
7-public is highly satisfied with snow and ice control
11-Good performance
12-78% satisfaction rate
14-People’s feedback varies, very little negative
17-50/50
18-Our surveys show that we provide good service.
20-93% approval of job being done
23-Mayors office surveys about all basic service functions
24- Expectations of the public are much higher than we expected. Overall they rated snow
and ice removal as a top priority to them and said that we did a good job.
26-We are doing a great job.
27. In your opinion, what three factors most account for your agency’s ability to improve
performance over the past few years?
a.
1- training
2- Evaluating past performance
B-15
3- Knowing where our performance is at
4-Workforce education
5-Better equipment
6-increased training, experience and usage of anti-icing
7-additional resources
8-anti-icing technology
9-Input from our statewide snow and ice committee
10-Anti-ice material
11-Set policy
12-Training
13-Computerized dispensing systems
14-n/a
15-Increase in the number of snow plows in our fleet
16-Greater amount of ice control materials stored under roof
17-n/a
18-Increased number of contract trucks
19-Technology
20-Empowering operators to adjust salt rates
21-Wet kits on trucks
22-Planning –time for personnel
23-good communication between police, fire, transportation, and PWD
24-Proactive operations
25-One person plowing
26-Wing plows
b.
1- staffing
2- planning for seasonal preparation
3- Having well defined performance objectives
4- Extensive RWIS
5- Better understanding of chemicals, particularly liquids
6-increased availability and usage of wing plows
7-new technology
8-improved material performance
9-operator training
10-Better weather forecasts
11-Proactive supervisory response
12- Training
13-Liquid applications
14-n/a
15-n/a
16-More qualified operators
17-n/a
18-Increased use of liquid de-icing materials to supplement salt.
19-Materials
20-Calibrating spreaders, pavement temperature guns in each truck
21-ground speed controlled spreaders
22-Joint cooperative to bulk purchase equipment
B-16
23-increased our parking fines and towing services
24-Materials Used
25-Better route management
26-Employee involvement
c.
1- dedication
2 N/A
3- Having a way to track performance
4-Innovative technology (anti-icing techniques, improved equipment, better forecasting)
5- Willingness to experiment
6-pride of the employees to their best
7-dedicated staff
8-Union commitment to do whatever it takes
9-Improvements in equipment design
10-n/a
11-Excellent training and retraining
12-Training
13-Increased staff
14-n/a
15-n/a
16-Better management personnel
17-n/a
18-Increased spotlight on performance measures.
19-Staff Education
20-zero velocity spreaders.
21-n/a
22-Joint cooperative to bulk purchase salt
23-Mayor’s support on clearing streets of vehicles blocking plows
24-Equipment improvements
25-Salt spreader calibration
26-new equipment technology
28. What have been the most significant barriers your agency has encountered in improving
performance?
1- the unknown
2- Cultural transformation-increased customer expectations
3- Communicating the department’s guidelines to all employees involved in snow removal
4-Education and overcoming existing paradigms
5- Getting over the hurdles utilizing new chemicals and more efficient rates
6-reduction of staff and resources
7-n/a
8-money
9-Lack of personnel and funding
10-Increase in traffic
11-County growth in numbers of miles of roads to obtained annually
12-Fleet turnover due to budgetary issues
13-Employee buy-in
B-17
14-$$$
15-We do not have snowfall on a regular basis so we don’t use the equipment enough to
continue to improve all the driver skills.
16-Weather forecasting and traffic
17-Inventory growth, customer expectations, and environmental costs.
18- SHA's shops and districts have grown comfortable in their current snow and ice control
operations. They have been very successful over the years and have shown limited interest in
changing their operations. The highest levels of the organization need to challenge personnel
to be more creative in order to get the job done with less materials and hired equipment.
19-Relating storm conditions/severity to costs
20-It’s not seen as an area to put research or new tech money
21-Funds shortages for purchasing. New Equipment: wet kits, gsc spreaders
22. Union issues
23-the amount of cars parked all over the city which has no off street parking and the
narrowness of the city streets
24-Careless drivers but budgets and staffing are always a factor
25-Union agreement
26-Budget/staffing/facility size
29. Excluding your own, which agencies around the world stand out as leaders in performance
measurement and/or management, in your opinion?
1 no opinion
2 Illinois DOT
3-N/A
4- PNS group and other Midwestern states
5- Colorado, Minnesota
6-unknown
7-Minnesota, Iowa, and Finland
8-States of Minnesota, Iowa, and Washington seem to be leaders.
9-Iowa DOT
10-n/a
11-APWA
12- Iowa DOT
13-Iowa DOT, Mn/DOT, Ohio DOT
14-n/a
15-no opinion
16-n/a
17-n/a
18-Iowa DOT and Colorado DOT
19-n/a
20-n/a
21-Iowa DOT
22-Ohio DOT
23- the feds
24-Washington DOT
25-no opinion
26-McHenry County, IL, MnDOT, Iowa DOT
B-18
APPENDIX C. AGENCIES USING PUBLIC SATISFACTION SURVEYS
State Agencies
Alaska DOT & PF
Caltrans
Colorado DOT
Illinois DOT
Indiana DOT
Iowa DOT
Kansas DOT
Minnesota DOT
New Mexico DOT
South Dakota DOT
Wisconsin DOT
Local Agencies
City of Cedar Rapids IA
City of Des Moines IA
City of Detroit MI
City of Edmonton AB
City of West Des Moines IA
El Paso County, CO
Washington County, MN
C-1
Other Countries
Swedish Road Administration

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