FRONTAL OCCUPANT SAFETY SIMULATIONS FOR COACH AND BUS PASSENGERS

C.G. Huijskens
M. Schrooten
P. de Coo
TNO-Automotive
The Netherlands

Paper no. 284

ABSTRACT well as to propose regulations and new or

improved cost effective crash test methods to
In the EU each year approximately 20000 decrease the injury risk for the bus occupants [1].
coaches of over 5000 kg are involved in This project was initiated because approximately
accidents that result in personal injuries. Each
30000 bus and coach occupants are injured every
year more than 30000 persons are seriously
year within Europe. Some 150 of these
injured in these accidents, and over 150
occupants of buses and coaches suffer fatal occupants suffer fatal injuries.
injuries. In contrast to other accident data, no
tendency for a significant reduction can be The task of TNO Automotive in the “Enhanced
found. Coach and Bus Occupant Safety” project
(ECBOS) is to suggest improvements to bus
Only three EC Regulations (Appendix A)
restraint systems and methods to evaluate safety
currently influence the structural and seat design
for buses and coaches. The general objective of of the bus passengers (M3 type). The work
the EC RTD project “Enhanced Coach and Bus reported here is concentrated on the effect of
Occupant Safety” is to generate new knowledge impact crashes of a type where the bus remains
that will allow further minimization of the in the upright position. This means that cases
incidence and cost of injuries caused by bus and were the bus is overturned, either directly, or
coach accidents. because it drives off the road and then overturns,
is not taken into consideration. This choice was
One of the main tasks in this project is to make a
detailed study of the occupant behavior by made because overturning accidents lead to
performing MADYMO simulations, so that the deformation of the passenger compartment and
injury causes in frontal impact can be to specific types of injury. Secondly, passengers
determined. A detailed bus and occupant model impacting each other or passengers being trapped
are used to investigate the following items: between the bus and the ground may cause
• The effect of different type of restraint injuries. This paper focuses on frontal impacts
systems, 2-point and 3-point seat belts. where the main interaction is between the
• The interaction between several restraint passenger and the restraint system, the forward
systems and different sizes of adult seat, a bulkhead or other solid object. Although
occupants, and, as a special case, of children this is a very limited subset of all injury causing
in (school) buses. loading conditions, it seems to be the only one
• Recommendations for improving ECE/R.80. for which the suitability and optimisation of
restraints systems makes sense.
These investigations have led to a virtual interior
assessment with multiple coach occupants, METHOD
including optimization of seat design parameters.
Simulation models like this can play an The main task of TNO Automotive in the
important role in identifying the benefit of new ECBOS project is to investigate how the
designs by application of new test methods and
implementation of restraint devices in M3 buses
regulations.
can reduce serious injury in frontal impacts.
INTRODUCTION
A number of sub-tasks have been identified:
Initiated within the European Vehicle Passive
Safety Network a consortium of 7 European 1. Develop a model of a typical ECE-R.80
Research Institutes and Universities was formed passenger seat as a basis for the optimisation
to investigate current bus and coach accidents as process.

Huijskens, 1
2. Develop a mathematical occupant model, Table 1.
using the standard seat, a variety of dummy ECE-R.80 requirements
sizes, and various restraint systems.
3. Optimise the protection offered to the Chest/Head < 1.6 m from SRP
occupant through study of key seat and displacement
restraint design parameters. This was done HIC < 500
for various dummy sizes. ThAC < 30g
FAC < 10 kN (all time)
4. Develop a generic model of a bus, which is
able to reproduce the kinematics of selected
Also a series of component tests was carried out
cases present in the ECBOS accident for measuring the stiffness and strength
database. characteristics of the main seat elements such as
5. Combine bus structure model and occupant the seatback, seatbase and seatpan/belt
model to ‘reconstruct’ a selected number of anchorage’s as required input variables for
cases from the ECBOS database. Verify that modelling (Figure 2).
the combined model is able to predict the
observed injuries.
6. Using the verified models, investigate how
improvements determined in sub-task 3
could be used to improve occupant safety in
real-world accident scenarios.

The occupant model is needed to obtain the

estimates of the injuries due to the loading of the
vehicle from change of velocities. The models
must be capable of describing the behaviour of
different dummy sizes and different restraint
systems like 2-point belt and 3 point belts.

The bus structural model is used to predict the

Figure 1 ECE-R.80 test
global kinematics of the bus in different impact
configurations, like full front wall impact, trailer
back barrier impact or offset impact. The model
consists of a multibody MADYMO model with
tire characteristics, wheel suspension
characteristics and global crash parameters of the
vehicle front

Finally, a first draft of new numerical test

methods as well as component- and full-scale
test methods is developed to improve the
existing ECE Regulation 80 and additionally to
propose new (simplified) test procedures for
frontal impacts.

SIMULATION MODELS-M3 VEHICLES Figure 2 Component test

A series of baseline sled tests was performed
Vehicle (Figure 3) and occupant (Figure 4)
according to the requirements of ECE-R.80
models have been created and validated for M3
(Table 1) in order to be able to evaluate the
buses in frontal impacts. The results of
dynamic performances of bus seat frames with
simulations performed in these tasks are used
both belted and unbelted occupants (Figure 1).
here to illustrate possible contacts.

Huijskens, 2
The injury parameters of the dummy models The data from each selected case was entered
indicate where injury criteria limits are being into a record of the ECBOS database:
exceeded.
1. vehicles involved.
Parametric studies have been carried out to 2. make and type of bus, mass, weight, etc.
investigate the influence on injury risk when
3. crash scenario, impact locations, and
certain key parameters, such as vehicle structure,
deformations of vehicle.
seat characteristics and stiffness are changed.
These results indicate areas of the vehicles that 4. passenger data, including sitting position,
could be improved. age, and sustained injuries.

In the development of the ECBOS database, each

selected case is investigated using the PC-
CRASH program. The program allows a
reconstruction of the accident and determines the
most likely initial velocities of the vehicle,
describes gross vehicle motion, orientation and
accelerations during the impact. In this way, the
dynamics and kinematics of the passenger
compartments of the vehicles involved were
determined.

These in-depth accident studies have generated

Figure 3 Vehicle model very valuable data. The data has been used to
improve and validate the simulation models.
However, the occupant injury data was limited.
Therefore it is not fully safe to summarise the
most important injury causing mechanisms
found within the studied accidents. Taking this
into account, a ‘sensitivity analysis’ was
performed to provide the most important
parameters for the head, neck, thorax and upper
leg injuries.

In-depth database analysis shows that single

accidents and overturning, which are combined
in the majority of the cases, cause the highest
risk for severe injuries. Frontal and rollover
accidents cause a similar proportion of fatalities
but rollover has a much higher risk (+ 42%) of
MAIS 3+ injury severity.

Figure 4 Dummy model INJURY MECHANISM-M3 VEHICLES

ACCIDENT DATABASE-M3 VEHICLES Frontal accident

A sensitivity analysis was performed to
The ECBOS database describes a number of determine the influence of a number of
serious accidents that have occurred on the roads parameters on the injury values.
in Europe. These accidents have taken place in
various European countries.

Huijskens, 3
It is concluded that for the upper part of the Simulations performed by the ECBOS partner
human body, the recliner stiffness has the biggest ‘POLITO’ [1] showed that for head injury
influence on the injury values. When the neither 2 point nor 3 point seatbelts prevent a
occupant is unbelted, the head-ashtray contact HIC value over 1000 for the occupant seated by
also has a large influence on the injury values the impacted side, the head always strikes the
(Figure 5) side window. If any seatbelt is used the injury
levels for occupants in the seats on the other,
Unbelted - Influence of variable on injury. non-impacted side of the bus, are always below
100.00%
90.00%
HIC
FNIC-T
the HIC limit. The real advantage of restraint use
80.00%
70.00%
FNIC-C here is the prevention of occupant movement and
NTE
60.00% NTF the loading of other occupants.
50.00% NCE
40.00% NCF
30.00% Thac
FAC-right
For the occupant in the inboard seat, near the
20.00%
10.00% FAC-right impacted side, a 2 point belt does not prevent the
0.00%
head injury mechanism of striking the side, but a
ht
ht

t
h

th
th

gh

gh

three point prevents this contact.

dt
t

ig
ig

id
id

id
i

ei

ei
He
e
l-W

-W

-W

-H
-H

-H
y-

y-

ee
ck
ec

cl

ee
ck
ra

ra
Re

ba

Kn
R

ba
ht

Kn
ht

at
As

at
As

Se

Se

In simulations an interesting injury mechanism

Figure 5 Influence on the injury values-unbelted of high load to the pelvis is indicated caused by
impact and contact with the armrest.
For the lower part of the body, the seat back to
knee contact stiffness is the most critical Intrusion of the roof and therefore direct contact
parameter. gives very serious and fatal injuries. There is no
doubt that if the roof structure does collapse
The kinematics of one occupant during the crash crush injuries will occur to those occupants in
can be affected by the presence of another the area of intrusion.
occupant. This was found to be especially
relevant when occupants are wearing a two-point Hand luggage causing injury is also a possibility
belt, as the occupants can introduce an additional during rollover accidents, although the falling of
loading to the recliner in front of them and luggage from overhead racks will also be likely
thereby influence the kinematics of the occupant in frontal accidents.
in front of them.
It is observed that the use of restraints would
From different research studies [4, 5, 6] it is prevent many serious injuries by preventing the
known that the most common mechanism of high degree of occupant interaction, interaction
fatal or serious injury in frontal accidents for M3 with the interior and ejection that occurs in
vehicles has been found to be direct intrusion. rollovers. The use of laminated glass may also
Many of the investigated cases feature large help to prevent contact with the ground and the
amounts of intrusion and structure deformation, ejection of occupants.
with impacts with trucks being a particular
problem. In these cases, it is very difficult to THE EFFECT AND CONSEQUENCES OF
suggest simple prevention, due to the collapse of
the bus structure in the area of the impact caused USING 2-POINT OR 3-POINT BELTS
by the high energy involved.
General
Rollover accident
It is observed that wearing a safety belt, 2 point
In order to be able to generate preliminary or 3 point is safer than wearing no belt. The main
recommendations for wearing 2 points or 3 advantage of wearing a belt in a bus or coach is
points seat belts in buses and coaches, it is preventing ejection during a rollover accident as
important to understand the difference in well as during a frontal accident.
occupant injury mechanism during frontal and
rollover accidents. Therefore the results of the Wearing a 2 point or 3 point seat belt in a bus or
analysis of rollovers accidents are also presented coach during a frontal impact poses a number of
in this chapter to prevent possible conflicting risks for the head and neck. A 95th percentile
consequences in the next chapters. dummy wearing 3 point belts was positioned in
the third seat row (Figure 4). Wearing a 2 point
seat belt show a higher risk of neck loading and

Huijskens, 4
head contacts during frontal impacts compared to in front of it. It is therefore absolutely necessary
a 3 point seat belt (Figure 6). to prevent it.
Additionally, wearing a 3 point seat belt reduces
the head injury of the behind-row passenger and With all tested belt configurations, it is also
the energy absorption capabilities of the seat observed, especially in the performed full scale
backrests in front of the occupant are not always sled tests, that the load path (belt-seat-floor)
enough to avoid injuries. allows too much deflection, which may result in
extensive head excursions. In the 3 point belt
2 Point Belt configurations little differences in the results are
200%
found when changing the seat pitch. Femur loads
180%
are reduced as the pitch increases. This is due to
160%

140%
the decreased interaction of the dummy with the
120%
forward seat.
1 Occupant
100%
3 Occupants

80% When the 3 point belt optimised parameters are

60%
applied to 2 point belt configurations the benefits
40%

20%
are no longer apparent. This is because increased
0%
seat back recliner stiffness needed for best
Chest 3ms HIC Nij TE Nij TF Nij CE Nij CF Femur Left Femur Right
results with the 3 point belt configurations.
2 point belt

3 Point Belt
Bulkhead
200%
To evaluate the effect of a bulkhead in front of
180%

160%
the dummy, a bulkhead model was used to
140% replace one row of seats in the occupant model.
120%

1 Occupant
The bulkhead structure was placed in front of the
100%

80%
3 Occupants
occupant, 72-cm forward of the rearmost point of
60%
the seat base structure. The geometry of the seat
40% model is shown in Figure 7. The model was
20%
evaluated with the 5th-percentile, 50th-percentile,
and 95th-percentile dummy models in the three
0%
Chest 3ms HIC Nij TE Nij TF Nij CE Nij CF Femur Left Femur Right

3 point belt point belt configuration.

Figure 6 Front and rear seat occupant influence

Integration of 3 point belt systems requires a

strong reinforcement of the seat backrest. The
consequence of the impact between an unbelted
occupant, and a seat backrest optimised for 3
point belts (more rigid) has also been evaluated.
It appears that a seat backrest optimised for 3
point belts will contribute to high injury levels in
the thorax, but lower on the head. Seats that are
not optimised for 3 point belts will do the
opposite. Therefore it is difficult to define any
advantage of low stiffness recliners against
more rigid ones, in the case of unbelted
occupant.
Figure 7 Bulkhead model
An extreme scenario will occur when an
unbelted 95th percentile occupant is seated at the
auxiliary seat, and a belted 50th percentile Compared to a configuration with a row of seats
occupant at the front seat. Under such loading it appears that the risk of injury to the femur is
conditions, the seat anchorage of the front seat decreased when the bulkhead configuration is
will experience extreme forces and may rupture with integrated three point belts.
under high severity crash. The rupture of just one
seat is likely to produce the rupture of the seats

Huijskens, 5
 RECOMMENDATIONS FOR IMPROVING

ECE-R.80
Children
It is clear that passive safety of bus and coach
Using a 3 point belt in combination with a child occupants will be improved if the following
below the age of 12 years requires specific recommendations could be implemented in ECE-
adaptations. An adult 3 point belt system may R.80:
not be able to protect child, and can even be
more harmful than protective when no specific • Combined dynamic test configuration;
adaptation is defined. It is therefore essential to • Worst-case seat/floor/sled combination;
evaluate the risk caused by not adapted three
• Requirements for child and small occupant
point belt systems, and the benefits of adaptable
restraint systems.
systems. Simulations have been carried out in
order to evaluate the benefit of three point belt
A sled test configuration could be two rows of
systems for children.
seats, the front with restrained passengers (50th
percentile dummies) and the rear with
In case no adaptation is made to the adult seat
unrestrained passengers (50th percentile
belt, the child occupant will have no choice but
dummies).
to wear a seat belt with a very high shoulder
attachment point. The main risk is that the child
Both the vehicle floor and the seat structure
occupant will experience major neck loads, due
affect the crash behaviour of the combination to
to direct contact between neck and belt. Also
be tested. To avoid having to tailor the bus seat
severe throat injuries, which cannot be predicted
of a certain seat manufacturer to the various bus
by simulation models, may result from such
and coach structures, the bus seats should be
loading conditions. In any case a 3 point belt
designed for a rigid floor structure that does not
must offer the possibility to lower the shoulder
absorb energy during impact. Tests performed on
attachment point. This adaptation is of interest
a combination of a rigid vehicle floor structure
not only for children, but also for small adults.
and seats specifically tailored to this structure are
applicable to all kind of different floor structures.
Simulations with 3 and 6 years old dummies
A special rigid floor structure and wall rail
wearing an adapted 3 point belt system (Figure
system should be defined for performing sled
8) have shown that good kinematics could be
tests according to ECE-R.80 [7].
obtained. However, the load on the thorax and
resulting injury criterions are high due to the fact
From the summary of ECE-R80, it is clear that
that the seat backrest and seat belt stiffness were
no interest is given to the necessary adaptation of
optimised for an adult 50th percentile occupant.
3 point belt systems to children or small
occupants. This probably is the main concern
related to this regulation, because wearing not
adapted 3 point belt systems can not be
considered as a solution for children. It seems
therefore necessary to update the ECE-R.80 with
respect to 3 point belt systems and the necessity
to either check the suitability of the belt system
for children or to limit the access to 3 point belts
for children.

The feasibility of implementing these

recommendations should be analysed.

CONCLUSIONS
Figure 8 Proposal for shoulder point
adaptation
Direct intrusion, seat ruptures due to the impact
of other passengers and ejection of the
In fact, 3 point seat belts adapted for children
passengers are the three main causes of injuries
require specific solutions to be designed for
during a bus or coach accident. Ejection out of
(school) buses and coaches.
the bus or coach through side window or

Huijskens, 6
windscreen is not only found in rollover Vehicles Safety and Reliability, 2-4
accidents, but also in frontal accidents and September 2002, Keszthely Hungary.
causes fatal injuries to the passengers. That’s
3. Mitsuishi H, Sukegawa Y, Matsukawa F,
why the use of seat belts, 2 points or 3 points, is
Japan Automobile Research Institute, Okano
strongly recommended for adult as well as for
S, Japan Automobile Manufactures
child passengers.
Association, Inc., “Crash-Safety
Performances of Bus Passengers”, 33rd
Additionally, it would be better to have all
Meeting of Bus and Coach Experts
children restrained during an accident, even with
International Conference on Vehicles Safety
a 2 point belt, than having them unrestrained, as
and Reliability, 2-4 September 2002,
the biggest risk to be injured is by ejection.
Keszthely Hungary.
Making 2 point or 3 point belt systems 4. Langwieder K, Gwehenberger J, Bende J,
obligatory in buses and coaches requires “Coaches and buses in the accident scene –
sufficient strength of the bus structure regarding Results of a study regarding passenger
the seat belt load path. The seat to floor protection”, 33rd Meeting of Bus and Coach
attachments should be sled tested on a rigid floor Experts International Conference on
at 50 km/h and 20g as this represents the worst Vehicles Safety and Reliability, 2-4
case. September 2002, Keszthely Hungary.
5. Páez Ayuso F.J, Aparicio Izquierdo F,
Finally, based on the best compromises between
Garcia Gracia A, INSIA-UPM, “Injury
wearing a 2 point or a 3 point belt system, the
mechanism analysis in accidents with buses
use of 3 point belt systems is recommended for
involveld”, 33rd Meeting of Bus and Coach
adult and child occupant passengers in buses and
Experts International Conference on
coaches.
Vehicles Safety and Reliability, 2-4
ACKNOWLEDGEMENTS
September 2002, Keszthely Hungary.
6. Mayrhofer E, Geigl B, Institut fuer
The work described in this paper form part of the Mechanik, Technische Universitaet Graz,
EC funded research programme ECBOS [1] Austria, Steffan H, Dr.Steffan Datentechnik,
conducted by the Safety Department of TNO Linz, Austria, “Enhanced Coach and Bus
Automotive. Occupant Safety through analysys of current
test procedures and real world accident
The help and co-operation of the Dutch bus incidents”, 33rd Meeting of Bus and Coach
manufacturer BOVA is gratefully acknowledged. Experts International Conference on
Vehicles Safety and Reliability, 2-4
The authors want to acknowledge also that the September 2002, Keszthely Hungary.
collaboration of the partners CIC (Cranfield
Impact Centre, UK), GDV (Gesamtverband der 7. Vogelsitze GmbH, Karlsruhe, Germany,
Deutschen Versicherungswirtschaft, DE) INSIA “Compliance wit European safety
(Universidad de Madrid, ES), POLITO requirements for coach seats”, 33rd Meeting
(Politecnico di Torino, IT), TUG (Technische of Bus and Coach Experts International
Universitaet Graz, AT) and VSRC Conference on Vehicles Safety and
(Loughborough University, UK) has been Reliability, 2-4 September 2002, Keszthely
essential to complete this paper. Hungary.
8. Crasas C, Ferrer I, IDIADA Automotive
REFERENCES Technology SA, “Application of accident
reconstruction in the assessment of the use
1. ECBOS, Enhanced Coach and Occupant of seat belts in buses”, European
Safety, EC project within Framework 5; Automotive Congress, Bratislava 18th-20th
Project No: 1999-RD.11130. June 2001.
2. Matolcsy M, Scientific Society of
Mechanical Engineers (GTE), “Frontal
collision of buses – problems, questions,
regulations”, 33rd Meeting of Bus and
Coach Experts International Conference on

Huijskens, 7
 APPENDIX A
RELATED REGULATIONS AND DIRECTIVES

Based on a resolution of the European Conference of Transport Ministers the co-ordination of technical
specifications for coaches was taken up by the ECE Working Party 29 in 1967 as their terms of reference.
In the process, a high level of safety was supposed to be obtained. Nine ECE Regulations dealing with
occupant safety requirements for buses and coaches resulted from these negotiations (Table 2).

ECE EC EC EC Scope Remarks

Regulation Directive Last New
Revision

14R05 76/115 96/38 M1-3, N1-3 Safety-belt anchorage’s

16R04 77/541 96/36 2000/3 M1-3, N1-3 Safety-belts and restraint systems
36R03 2001/85 M2, M3 Uniform provisions concerning the
(> 22+1) approval of large passenger vehicles
w.r.t. their general construction
43 92/22 M1-3, N1-3 Safety glazing materials
52R01 2001/85 M2, M3 Uniform provisions concerning the
Single-deck construction of small-capacity Public
(< 22+1) Service Vehicles
66R - - 2001/85 M2, M3 Uniform provisions concerning the
(> 22+1) approval of large passenger vehicles
w.r.t. the strength of their
superstructure
80/R01 74/408 96/37 M2, M3 Seats of large passenger vehicles,
their anchorage’s and installation of
seats
107R 2001/85 M2, M3 Uniform Provisions concerning the
Double-deck approval of double-deck large
(> 22+1) passenger vehicles w.r.t. their general
construction
Table 2 Overview of existing Regulations and comparable Directives.

These ECE Regulations came into force between 1976 and 1989. The application of some of these ECE
Regulations is still not obligatory in all countries within the EC. For the passive safety of single decked
touring cars (M2, M3) there are only three ECE Regulations today which are of importance: Regulation
No. 80 (Seats of large passenger vehicles, their anchorage’s and installation), Regulation 14 (Safety-belt
anchorage’s) and Regulation 16 (Safety-belts and restraint systems). Although, these Regulations (and
comparable EC Directives) are not compulsory in all European countries, but they are taken into account
by most bus manufacturers in the development of new bus and coach model types and by most authorities
for approval.

Huijskens, 8