Journal of KONES Powertrain and Transport, Vol. 20, No. 1 2013
ISSN: 1231-4005
e-ISSN: 2354-0133
DOI: 10.5604/12314005.1136231
CHILD SAFETY IN A CAR DURING A FRONTAL COLLISION
Andrzej ĩuchowski
Military University of Technology, Faculty of Mechanical Engineering
Institute of Motor Vehicles and Transportation
Gen. S. Kaliskiego Street 2, 00-908 Warsaw, Poland
tel.: +48 22 6837454, fax: +48 22 6839230
e-mail: azuchowski@wat.edu.pl
Abstract
Road accidents with participation of the children are of particular nature. According to surveys, the older
children at the age of 8-12 years are often transported in cars in a wrong way. They are usually fastened with seat
belts designed for adult passengers. In order to make the seat belts safe for a child, they should be used together with
a booster cushion or a high back booster seat.
This paper presents the test results and analysis of dynamic loads affecting a child in the back seat of a passenger
car during a frontal collision. The car body was accelerated to a speed of about 48 km/h and then it was suddenly
stopped in a time of about 100 ms. The analysis was carried out on the basis of results of experimental tests performed
at the collision station in the Automotive Industry Institute in Warsaw. The attention was paid to the influence
of protection device type on the P10 dummy, representing a child at the age of about 10 years. Three child protection
options during a collision were considered: the dummy was sitting directly in a car seat, on a booster cushion and on
a seat. The dummy was fastened with standard seat belts. Evaluation of the protection device influence on the dummy
loads was given in association with the operation of the seat belts. During evaluation, the indexes of biomechanical
resistance of a human body to the impact load effects were used. They made the basis for the evaluation of the child
injury severity risk.
Keywords: road transport, vehicle safety, crash tests, child safety, booster seat
1. Road accidents involving children and their effects
Road accidents are still the most frequent external cause of children’s death [10]. In Poland,
2296 children at the age of 0-14 years died in road accidents during the period of 2000-2012. Their
number is systematically decreasing, from 265 in 2000 to 89 in 2012 [6, 15]. Considering the
changing population of the children at the age of 0-14 years, the fatal injury index per 1 million
children amounted to 36.3 in 2000 and it amounted to 15.3 in 2012 and it confirms a significant
improvement of the children safety in the road traffic.
Children at the age of up to 14 years mostly die on the roads as the passengers of the passenger
cars (about 45% children were killed in the road accidents in EU-15 during 2003-2009 [4, 6],
about 52% in Poland during 2010-2012 [15]). In Poland, children at the age of up 14 years make
about 8% of victims and 15% injured ones among passengers of the passenger cars [15].
Detailed data on the road accidents involving children is not available in Poland (including
places occupied in a car, type of collision, child restrain method). On the basis of the data on the
road accidents in the USA [1] Tab. 1 gives the index values specifying a number of severely
injured or killed (MAIS3+) per 1000 children occupying the rear seats. It only includes accidents
with restrained children. This data confirm that in most accidents older children are worse
restrained than the younger ones. Frontal collisions are particularly dangerous for the children at
the age of 9-12 years.
The paper [3] finds out, on the basis of the road accident data from the period of 2000-2005,
involving children at the age of up to 10 years, transported in cars, that about 61% of all fatal
injuries and severe injuries were among the children at the age of 5 to 10 years, 32% among the
A. ĩuchowski
children at the age of 1-4 years and the remaining 7% at the age of below 1 year. More victims
among older children probably results from that fact that they travel by car more frequently but it
can also be caused by improper children restraint in a car.
Tab. 1. The number of serious injured or killed (MAIS3+) per 1000 restrained children, according to age and crash
type (USA, 1991-2005) [1]
Crash type
Frontal
Side
Rear
Rollover
4 to 5 years
2.8
6.5
7.1
3.1
6 to 8 years
4.4
4.1
1.2
6.7
9 to 12 years
16.0
7.7
0.6
7.1
The above data on the road accidents results in the following findings:
majority (51-54%) of children at the age of 0-14 years killed in the road accidents were the
passengers of the passenger cars,
older children at the age of about 10 years are in the group of the highest risk,
frontal collisions are the most dangerous among various types of accidents, they make
10-12% of road accidents but they generate as much as 19-20% fatal injuries [15].
Considering the above, the following part of the paper pays attention to the risk of injury for
a child at the age of about 10 years, occupying the rear seat of a car during a frontal collision.
2. Children injuries in cars during road accidents
A child in a car should be transported in a restrain seat adjusted to its figure (weight, height).
Seats for small children are fixed to a car seat and are equipped with their own belts for fastening
a child. Older children usually use the booster seats designed to adjust the seat belt position on
a child’s body (Fig. 1).
Fig. 1. Booster cushion and high back booster seat [11]
Data on the child injury in the road accidents in Poland is rarely published. It usually refers to
a small number of children at the age of about 10 years and does not include information on a type
of collision and child restraint method while driving [7, 12]. These both papers state that head and
neck injuries were the most frequent child injuries.
More detailed statistical data on child injuries during road accidents are available in foreign
literature. The paper [5] specifies that during a frontal collision the risk of head and stomach
injuries (AIS2+) for the children transported on a booster cushion and fastened with seat belts
were 20-30% lower than for the children without such a booster cushion. Tab. 2 gives the results
of analysis of the children injuries during a frontal collision from the European CREST (Child
396
Child Safety in a Car During a Frontal Collision
Restraint Systems for Cars) database [5]. A number of the least serious injuries (AIS3+) of
individual parts of the body were associated with various child restraint methods: rearward facing,
forward facing devices, booster and safety belt, safety belt only. Serious injuries of younger
children transported in car seats mostly refer to the head and the neck. Children seating on
a booster cushion and fastened with seat belts also suffered from chest and stomach injuries.
Children fastened only in a seat belt mostly suffered from severe stomach and chest injuries.
Children in all groups often suffered from limbs fractures.
Tab. 2. CREST accident database: Rearward facing / Forward facing devices / Booster and safety belt / Adult safety
belt only (frontal impact, AIS3+) [5]
Number
of children
Head
Neck
Chest
Abdomen
Limbs (*- fractures )
Rearward facing
systems
31
5
0
0
0
4*
Forward facing
systems
144
16
10
6
3
20*
Booster + seat belt
108
7
11
9
9
25*
Adult safety belt
only
148
8
6
18
27
38*
Similar results of the children injury analysis are described in the papers [2, 13]. The paper
[13], on the basis of the data on the road accidents in Europe from the period of 1998 to 2008,
involving children at the age of 6-12, finds that over 70% of 207 children were transported only in
the seat belts and 15% without any restraint. None of the children seating on the high back booster
seats suffered from AIS2+ injuries. This injures occurred in:
- 8.7% children restrained in adult seat belt only (13 of 149),
- 18.7% children restrained in booster cushions (3 of 16),
- 24.1% unrestrained children (7 of 29).
Restrained children mostly suffered from stomach, head and upper extremity injuries. In case of
side collisions (79 children at the age of 6-12 years), passengers mostly suffered from head injuries.
Data given on Fig. 2 refer to the accidents in Australia in the period of 2003-2005 [2], involving
152 children at the age of 2-8 years. There were 15 to 27 children at that age. Frontal collisions made
about 50% of accidents and side collisions made 28%. Seat belts were the only restraint for about 60%
of the children at the age of 4-6 years and over 90% of the children at the age of 7-8 years.
a)
b)
Fig. 2. Relationship between distribution of injury severity, age and restraint (a) and distribution of AIS2+ injuries
for different restraint types (b) [2]
Data from the Fig. 2 confirm that the children restrained with CRS/Booster more rarely
suffered from injuries than the ones restrained only with a seat belt. Child’s head injury was the
most frequent one. Younger children suffered from severe internal injuries of the abdominal cavity
in frontal collisions were identified, where a lap belt was placed too high and it moved to
a stomach resulting in severe injuries [2].
397
A. ĩuchowski
3. Experimental tests of the child dummy loads during frontal collisions
Previous papers [16, 17] present comparative evaluation of loads that affect the passengers
occupying the front and the rear seats in a car during a frontal collision. It was found that loads of
the dummies in the back seats, including the children at the age of 10 years, are usually several
times higher than for the dummies in the front seats.
The evaluation of the influence of restraint equipment on a child load condition during a frontal
collision is presented below. The attention was focused on the dummy’s head and torso loads P10,
representing a child at the age of about 10 years. Evaluation of the protection device influence on
the dummy loads was given in association with the operation of the seat belts. The analysis was
carried out on the basis of results of experimental tests performed at the collision station AB-554
in the Automotive Industry Institute in Warsaw. The car body was accelerated to a speed of about
48 km/h and then it was suddenly stopped in a time of about 100 ms. The course of the car body
delay was in accordance with requirements defined in the Regulations ECE R44, which refers to
the tests on the child restraint devices. The maximum car body delay amounted to about 22 g. In
the next crash tests, high car body delay repetition was obtained which is crucial for the analysis
carried out further on.
The P10 dummy was seating on the left side of the back seat and was fastened with standard
seat belts that were replaced with new ones after each crash test. During the next crash tests
(marked as B, P, F), the P10 dummy was restrained in different ways (Fig. 3):
B – restrained in adult seat belt only,
P – restrained in booster cushions (for children of 22 to 36 kg),
F – restrained in hi-back booster seat (for children of 15 to 36 kg).
Values measured during the crash tests included:
- forces extending the seat belt band in lap and shoulder sections,
- dummy’s head and torso acceleration in three mutually parallel directions.
Crash tests were filmed by means of video cameras designed for fast shooting (1000 frames per
seconds).
Test B
Test P
Test F
Fig. 3. P10 dummy in a car body
Executions of forces extending the seat belt band were given in Fig. 4. Executions of the force
in the shoulder belt (SB) and in the lap belt (LB) were compared separately. The tests B, P and F
show various force increase paste in the shoulder belt, while the maximum values of that force and
time of occurrence are similar. In case of the force in the lap belt, courses are significantly
different: the maximum values and time of occurrence are different. On the basis of the frame-byframe analysis of the videos, it was found that torso displacement in the direction of driving is the
highest when the force in the shoulder belt reaches the maximum value.
In the B test, the dummy is located directly on the back seat, the lap belt band is placed too
high over the thighs. The video shows that hip displacement against the back seat is significant
during the collision. The lap belt slides from the dummy’s thighs and hips to the stomach and
398
Child Safety in a Car During a Frontal Collision
further on under the ribs. Internal organs in the stomach and the chest are subject to severe injuries.
That effect, the so-called submarining, is visible as a temporary decrease of the force in the lap belt
during 85-95 ms (Fig. 4). In the F test, the dummy is placed on the high back booster seat and it is
placed more to the front against the backrest of the seat compared to the B and P tests (Fig. 3).
That dummy position resulted in a rapid force increase in the lap belt.
600
600
B / SB
P / SB
400
B / LB
500
Force [daN]
Force [daN]
500
F / SB
300
200
P / LB
400
F / LB
300
200
100
100
0
0
0.00
0.03
0.06
0.09
0.12 [s]
0.00
0.03
0.06
0.09
0.12 [s]
Fig. 4. A course of forces in seat belt (SB – shoulder belt, LB- lap belt)
The resultant values of the dummy’s head and torso acceleration are specified on Fig. 5. Their
maximum values are several times higher than a car delay (22 g), which is caused by delay in the
influence of the seat belt on the dummy (about 30 ms). The maximum values of the resultant torso
acceleration are the highest in the test F (high back booster seat), where the force in the lap belt is
much higher than in the tests B and P. The maximum values of the resultant head acceleration are
the highest in the test B.
a)
[g]
[g]
90
90
B
P
F
Vehicle
60
60
b)
B
P
F
Vehicle
30
30
0
0
0.00
0.03
0.06
0.09
0.12 [s]
0.00
0.03
0.06
0.09
0.12 [s]
Fig. 5. The resultant values of the dummy’s head (a) and torso (b) acceleration in the tests B, P and F
The influence of the restraint device on the dummy motion is much more complex than the
above description. More detailed analysis of the kinematics and the dummy’s loads in the tests B,
P and F will make a subject of a separate paper. Below, the attention is paid to the effects of the
restraint device change on dynamic loads of the P10 dummy. For that purpose, the results of the
dummy’s head and torso acceleration measurement results were used to calculate two injury
indexes:
- Head Injury Criterion (HIC36), calculated at time up to 36 ms,
- resultant maximum chest acceleration (CAcc), at time at least 3 ms.
The results of the calculations for HIC36 and CAcc indexes are given on Fig. 6. In the test B,
where the P10 dummy was seating directly on the rear seat, the value of the HIC36 index is several
times higher than in the tests P and F. The highest torso acceleration occurred in the test F (63 g).
It is higher by 43% than in the test P and higher by 18% in the test B and results from effective
operation of the lap belt.
399
A. ĩuchowski
2000
80
1672
1000
603
500
CAcc, g
HIC36
1500
483
60
63
52
44
40
20
0
0
B
P
B
F
P
F
Fig. 6. The results of the calculations for HIC36 and CAcc
The methodology described in the paper [16] was applied to evaluate the risk of the dummy’s
injuries. It is based on relations between the index values HIC36 and CAcc and the risk of injuries
described by means of a Abbreviated Injury Scale (AIS). A probability of severe head and chest
injury at the level of AIS4+ is described by the functions [8, 9]:
Phead ( AIS 4) {1 exp[5.02 0.00351 ˜ HIC36 ]}1 ,
(1)
Pchest ( AIS 4 ) {1 exp[5.55 0.0693 ˜ C Acc ]}1 .
(2)
The risk of death or disability of a person who suffered many injuries is higher than if the
injuries apply only to one part of a human body. It includes Pcomb (combined injury probability
criterion), usually expressed in percentage values. Including (1) and (2) we have [8]:
Pcomb 1 (1 Phead ) ˜ (1 Pchest )
Phead Pchest Phead ˜ Pchest .
(3)
The results of the calculations are given on Tab. 3. The values of the risk index Pcomb are the
highest in the test B and it is caused by a significant head load. In the test P, the risk of injuries on
the level of AIS4+ is 6-times smaller and it is 3 times smaller in the test F.
Tab. 3. A risk of severe head and chest injury at the level of AIS4+ (P10 dummy)
Test
HIC36
CAcc
Phead (AIS4+)
Pchest (AIS4+)
Pcomb (AIS4+)
B
1672
52 g
70%
12%
74%
P
603
44 g
5%
8%
12%
F
483
63 g
3%
23%
26%
Obtained results of injury risk calculations were compared to the results previously described
in the paper [16]. For that purpose, Fig. 7 associates the index values HIC36 and CAcc from
the tests B, P and F with values obtained during the analysis of the tests performed by NHTSA
[14, 16]. They referred to the child dummy at the age of about 10 years (Hybrid III, 10YO) loads
during frontal collision of various passenger cars into a rigid barrier at the speed of 56 km/h.
A 10YO child dummy was restrained in high back booster seat. Lines on a diagram indicating
injury risk limits AIS4+: 10, 20, 35 and 45% respectively.
Results obtained in the test B, P and F confirm serious influence of the applied type of the child
restraint device on the load level and injury risk. The P10 dummy head load in the tests P and F
(48 km/h) is lower compared to 10YO dummies (56 km/h), as probably affected by the collision
velocity value. In the test B, the load of the P10 dummy seating directly on the back seat is much
higher than for the majority of the 10YO dummies seating in the high back booster seats, despite
lower collision velocity.
400
Child Safety in a Car During a Frontal Collision
2250
2000
B
1750
HIC36
1500
1250
1000
750
F
P
500
250
10%
0
30
40
50
35%
20%
60
70
45%
80
CAcc [g]
Fig. 7. Association of dummy head and chest loads with AIS4+ injury risk (lines), B, P , F – P10 dummy (48 km/h);
Ÿ – Hybrid III-10YO (56 km/h)
4. Summary
The number of children killed in road accidents systematically decreases. However, road
accidents still remain the major external cause of death for the children at the age of 7-12 years.
Car passengers make over a half of the children at the age of 0-14 years killed on the roads.
Older children, 7-12 years old, are usually transported on a booster seat without a backrest or
directly on the back seat. During frontal collisions, they are subject to severe head and stomach
injuries, particularly in the second case [13].
The restraint device type determines the initial dummy position against the rear seat, its way of
displacing against the seat and thus the operation of the seat belt. The laboratory test results given
in this paper confirm very high dynamic loads affecting the P10 dummy, seating directly on the
rear seat during a frontal collision (test B). Despite the fastened seat belt, the severe head injury
risk (AIS4+) amounts to as much as 70%. When the dummy was seating on the booster seat the
head injury risk amounted to just 3-5% (tests P and F).
The scope of the child injury risk analysis given in the paper was limited by the P10 dummy
measurement abilities and did not include stomach and neck loads that often occur for the children
involved in road accidents.
Acknowledgement
This work has been prepared according to the project No. N N509 559 640, financed from the
funds of the National Centre of Science, Poland.
References
[1] Bidez, M. W., Burke, D. S., King, D., Mergl, K. M., Meyer, S. E., A Critical Safety Need for
Children Ages 9 to 12 in the Rear Seat, 5th International Conference – Protection of Children
in Cars, Munich, Germany 2007.
[2] Brown, J., Bilston, L., McCaskill, M., Henderson, M., Identification of Injury Mechanisms for
Child Occupants aged 2-8 in Motor Vehicle Accidents, Final Project Report for the Motor
Accidents Authority of New South Wales, Australia 2005.
[3] Charlton, J., Koppel, S., Fitzharris, M., Congiu, M., Fildes, B., Factors that Influence
Children's Booster Seat Use, Monash University Accident Research Centre, Report No. 250,
Australia 2006.
[4] Childhood Road Safety, The European Child Safety Alliance – EuroSafe, Consumer Safety
Institute in the Netherlands 2007.
401
A. ĩuchowski
[5] European Enhanced Vehicle-Safety Committee, EEVC Working Group 18 Report Child
Safety – February 2006, (www.eevc.org).
[6] European Road Safety Observatory, Traffic Safety Basic Facts 2011: Children (Aged < 15),
2011, (http://ec.europa.eu/transport/road_safety/specialist/index_en.htm).
[7] Goniewicz, M., Wnuk, T., Ostrowski, M., Nogalski, A., Kulesza, Z., Wypadki komunikacyjne
jako przyczyny obraĪeĔ u dzieci, Zdrowie Publiczne 116(1), pp. 158-161, 2006.
[8] Hackney, J. R., Kahane, C. J., The New Car Assessment Program: Five Star Rating System
and Vehicle Safety Performance Characteristics, SAE Technical Paper 950888, 1995.
[9] Hong, S., Park, Ch., Morgan, R., Kan, C., Park, S., Bae, H., A Study of the Rear Seat
Occupant Safety using a 10-Year-Old Child Dummy in the New Car Assessment Program,
SAE Technical Paper 2008-01-0511, 2008.
[10] Kuáaga, Z., Litwin, M., Wójcik, P., Jakubowska-Winecka, A., Grajda, A., Gurzkowska, B.,
Napieralska, E., Barwicka, K., RóĪdĪyĔska, A., WiĞniewski, T., Current trends in external
causes of children and adolescent mortality in Poland. Problemy Higieny i Epidemiologii
90(3), pp. 332-341, 2009.
[11] Materiaáy informacyjne firmy Graco, (http://www.gracobaby.pl).
[12] StĊpniewski, W., Rydzewska-Dudek, M., Janica, J., Zaáuski, J., Niemcunowicz-Janica, A.,
PtaszyĔska-Sarosiek, I., Dopieraáa, T., Okáota, M., Szeremeta, M., Analiza ofiar wypadków
drogowych w wieku 0–18 lat w materiale sekcyjnym zakáadu medycyny sądowej Akademii
Medycznej w Biaáymstoku w latach 1997-2006, Roczniki Pomorskiej Akademii Medycznej
w Szczecinie, No. 53-2, pp. 50-54, Szczecin 2007.
[13] Visvikis, C., Pitcher, M, Girard, B., Longton, A., Hynd, M., Literature review, accident
analysis and injury mechanisms, Final Project Report 218744, EPOCh Consortium, 2009.
[14] www.nhtsa.gov.
[15] Wypadki drogowe – Raporty roczne, (http://dlakierowcow.policja.pl).
[16] ĩuchowski, A., Risk of Injury for the Front and Rear Seat Passengers of the Passenger Cars
in Frontal Impact, Journal of KONES Powertrain and Transport, Vol. 19, No. 3, pp. 507-518,
2012.
[17] ĩuchowski, A., Prochowski, L., Analysis of the Influence of the Seat Taken in a Car on the
Dynamic Load Affecting People During the Obstacle Impact by the Passenger Car,
Automotive safety problems, Vol. 1., Ed. by StaĔczyk L.T., Wydawnictwo Politechniki
ĝwiĊtokrzyskiej, pp. 113-126, Kielce, Poland 2012.
402