The work described in this report has been carried out on behalf of the UK Department for Transport by TRL Limited. The aim of the work was to assess the safety of wheelchair users when being transported on all M category vehicles in comparison with travellers seated in conventional seats (fitted with headrests). In cases where the safety of the wheelchair user was lower than that of other passengers, or considered unacceptable for other reasons, modifications were assessed.
The approach to the work involved a programme of numerical simulation followed by an extensive programme of testing involving 37 individual sled impact tests. In addition, the safety of passengers under normal transit conditions was addressed.
The work found that the heads and necks of wheelchair users were particularly vulnerable but that this could be addressed through the use of a head and back restraint. However, such a restraint should meet the requirements of ECE Regulation 17 for strength and energy absorption and the wheelchair should fit well up against the head and back restraint for maximum benefit.
Further recommendations from the work were that an upper anchorage location for diagonal restraints is preferable to a floor mounted location and that the restraint anchorages should meet more rigorous strength requirements than are required at present. A protected space envelope for forward facing wheelchair passengers is also recommended.
Under normal transit conditions a vertical stanchion is preferable to a horizontal bar in terms of preventing excessive movement of the wheelchair.
|
EC |
European Commission |
|
ECE |
Economic Commission for Europe |
|
DfT |
Department for Transport |
|
TRL |
Transport Research Laboratory |
|
MDA |
Medical Devices Agency |
|
PSV |
Public Service Vehicle |
|
DDA |
Disability Discrimination Act |
|
C&U |
Road Vehicles (Construction and Use) Regulations 1986 |
|
PSVAR |
Public Service Vehicles Accessibility Regulations 2000 |
|
ISO |
International Standards Organisation |
|
EC WVTA |
EC Whole Vehicle Type Approval |
|
DPTAC |
Disabled Persons Transport Advisory Committee (- the Government's statutory adviser on the transport needs of disabled people) |
|
COST |
Co-operation in the field of Scientific and Technical Research |
|
M1 Vehicles |
Vehicles with = 8 seats in addition to the driver's seat |
|
M2 Vehicles |
Vehicles with > 8 seats in addition to the driver's seat and a maximum mass = 5 tonnes |
|
M3 Vehicles |
Vehicles with > 8 seats in addition to the driver's seat and a maximum mass > 5 tonnes |
|
MPV |
Multi Purpose vehicle |
|
EuroNCAP |
European New Car Assessment Programme |
|
NIC |
Neck Injury Criteria |
|
DRTF |
TRL's Dynamic Restraint Test Facility |
|
FE |
Finite Elements |
|
MADYMO |
Proprietary 'Multi-body' Numerical Modelling Code |
|
ATF |
Aluminium Track Fittings |
|
RAGB |
Road Accidents in Great Britain Report |
Over recent years a number of legislative tools and codes of practice have been put in place to provide wheelchair users with greater access and freedom of use of public transport. Such regulations range from guidelines issued by national Governments, with the UK Government taking a lead role, to full EC directives. While these positive steps have achieved the aim of providing a greater choice and freedom of transport use to wheelchair users, the issue of safety in the event of an accident has not been rigorously assessed in a consistent manner across the various categories of road vehicles available to this group of travellers.
This project, commissioned by the UK Department for Transport, aimed to address the safety of adult wheelchair users in M1, M2 and M3 vehicles, i.e. private vehicles, taxis, minibuses, coaches and urban buses. The objective was to make recommendations for requirements on these categories of vehicles that would provide wheelchair users with at least an equivalent level of protection as a passenger seated in a conventional seat (fitted with a headrest) in the event of an accident. In addition, the security of carriage of a wheelchair user in an urban bus under normal operating conditions was also investigated.
The project tackled these issues firstly through a programme of numerical simulation, validated against a limited number of physical tests, the results of which helped to define a wide ranging testing programme. Initial work reviewing suitable test conditions indicated that the scope of vehicles could be addressed by examining 4 sets of conditions:
The protection provided for passengers was tested using conventional automotive crash test dummies, and the risk of injury assessed using the usual injury criteria derived from the dummy outputs. In each case a conventionally seated passenger configuration was tested to determine a comparable level of protection for the wheelchair seated occupant.
M1 and M2 vehicles were able to be considered together as previous research has shown that the same deceleration pulse is appropriate for the majority of both categories.
The modelling work indicated that the most influential parameters on the safety of wheelchair passengers are the location of the diagonal belt upper anchorage (i.e. upper location or floor level), the presence or otherwise of a head and back restraint and the closeness of fit between the wheelchair and the head and back restraint if fitted.
For forward facing occupants in M1 and M2 category vehicles it was apparent that some injury criteria such as head displacement and lumber spine compression were better for the wheelchair occupant than the conventionally seated occupant, however neck loads in particular were higher. The addition of a head and back restraint was found to improve the situation significantly, although the presence of a gap between the head and back restraint and the wheelchair had a detrimental effect. Any such head and back restraint should be compliant with the strength and energy absorption requirements of ECE Regulation 17. In general, an upper anchorage was preferable to a floor mounted anchorage.
Rear facing wheelchair passengers in M1 and M2 vehicles were found to be greatly more at risk than equivalent vehicle seated passengers, particularly in terms of neck and spine loads, the situation being worse still for both smaller and larger than average persons. Again, the situation was mitigated through use of a head and back restraint compliant with ECE Regulation 17, assuming a minimal gap between the wheelchair and the head and back restraint and a minimum horizontal strength requirement of 100kN.
The situation for forward facing passengers in M3 vehicles was similar to that for M1 and M2 vehicles, and the findings were also similar in that a head and back restraint was of benefit (compliant with ECE Regulation 17) with no gap and an upper belt anchorage.
Rear facing wheelchair passengers in M3 vehicles fitted with a back restraint not intended to provide crash protection, were found to be subject to unacceptably high head accelerations. The use of a head and back restraint compliant with Regulation 17 resolved the issue.
In all cases the anchorage loads were recorded and recommendations made for requirements on the anchorage strength in vehicles of each category. Likewise, occupant space requirements were derived from the dummy excursions for forward facing occupants.
The normal transit tests revealed that a vertical stanchion provides a better restraint on excessive wheelchair movement than does a horizontal bar. However, the tests only used a single type of wheelchair and hence any conclusions should consider the potential interaction of these systems with other wheelchair types.
The findings from this work have been developed into a set of recommendations for each category of vehicle which may form the basis for changes to regulations at the discretion of DfT.
The work described in this report was carried out by TRL Limited under contract to the Mobility and Inclusion Unit of the DfT. The test work took place over a 12 month period from 2001 to 2002 and examined, using both numerical simulation and physical testing, the requirements that should be made of a vehicle such that wheelchair users might be transported without being placed at an unreasonable risk of injury in the event of an accident. The work was overseen by a Steering Committee that comprised representatives from DfT, TRL, a wheelchair tie-down and occupant restraint manufacturer, PSV operator and the Medical Devices Agency (MDA).
While there are a number of regulations and codes of practice that impinge upon the travelling wheelchair user, a definitive programme of work specifically addressing the situation has not been undertaken previously. This work therefore aims to address this deficiency and provide the necessary background understanding as to the safety of wheelchair users in the event of an impact. In making recommendations on the basis of the work carried out, the existing requirements must be considered and any conclusions made in the context of the current regulatory framework.
In recent years, there have been significant advances in the availability of accessible transport. Accessibility regulations drafted under the Disability Discrimination Act (1995) will ultimately ensure that all forms of land-based public transport are accessible to wheelchair users and will require operators to provide for people who cannot transfer from their wheelchair into a vehicle seat.
The Road Traffic Act 1988 and the Public Passenger Vehicles Act 1981 (as amended by the Road Traffic Act 1991) provide the framework for most of the important provisions relating to the use of motor vehicles on roads in the United Kingdom. Section 40A of the 1988 Act states that a person is guilty of an offence if a danger or injury is caused to any person because of, for example, the manner in which passengers are carried or the load is secured in a vehicle. The Road Vehicles (Construction and Use) Regulations 1986 (C&U), as amended, govern the construction, equipment, maintenance and use of road vehicles. In addition, buses and coaches that are public service vehicles must comply with the Public Service Vehicles (Conditions of Fitness, Equipment, Use and Certification) Regulations 1981. Neither of these regulations provides requirements for the carriage of passengers in wheelchairs. However Regulation 100 of C&U requires that all passengers be carried in a manner such that 'no danger is caused to any person'. This generally refers to latent defects or, for example, unsecured loads. However it could be interpreted that if a wheelchair is not secured and an incident occurs as a result, an offence has been committed.
The Disability Discrimination Act, DDA (1995) aims to tackle discrimination against disabled persons. Part V of the act gives Ministers the powers to establish accessibility regulations that will ensure it is possible for wheelchair users to be carried in safety in land-based public transport whilst remaining in their wheelchair. These powers were first exercised for road vehicles in the form of the Public Service Vehicles Accessibility Regulations 2000 (PSVAR). These require regulated buses and coaches of more than 22 passengers on local or scheduled services to be wheelchair accessible. They list a number of requirements related to the safety of passengers in wheelchairs, including the direction in which they must face in the vehicle, the need for active and passive restraint systems and how these and their anchorages should be tested. The regulations initially apply to new vehicles only, but will apply to all regulated vehicles within 20 years.
Accessibility regulations under PSVAR make requirements of the vehicle only, enabling it to be certified for use. Manufacturers of wheelchairs have been aware of the transport needs of their equipment for some time. This area was recognised in legislation in the form of the Medical Device Regulations (1994), recently updated by the 2002 Regulations. These regulations, in accordance with the Consumer Protection Act (1987), require manufacturers to conduct a full risk analysis process to support the CE marking of their products. As part of this risk assessment, an international standard ISO 7176/19 for the impact testing of wheelchairs is given as supporting evidence of the suitability of a wheelchair to travel in a vehicle. This test is essentially a product test - it tests whether the wheelchair is able to take the loads imposed on it in the event of a road traffic accident - although excursion limits are placed on the dummy. However, it could be argued that, given that instrumented dummies are not used in these tests, it is not known whether the occupant would survive the incident as survivability is a compromise between excursion and accelerations to the body.
A number of wheelchair tie-down and occupant restraint systems are available on the market and these can be tested to International Standard ISO 10542. The test is similar to ISO 7176/19 and uses a 'surrogate' wheelchair in each test, defined as a "rigid, reusable" wheelchair that simulates a powered wheelchair for the purposes of testing wheelchair tie-down and occupant restraint systems.
The majority of M1 vehicles (see section 2.1.2) are subject to EC Whole Vehicle Type Approval (EC WVTA) whereby a vehicle must comply with a number of EC Directives. There are no directives covering provisions for the carriage of a wheelchair in these vehicles. Proposals are also in place to extend EC WVTA to other vehicle types including M2 and M3 vehicles. One of the EC Directives for M2 and M3 vehicles is Directive 2001/85/EC with provisions for the carriage of wheelchairs of which some are based on PSVAR.
Apart from legislation, there is also a long-standing Department for Transport code of practice (VSE 87/1) covering the safety of passengers in wheelchairs on buses. Its application is now limited to buses and coaches not covered by the Public Services Vehicles Accessibility Regulations (2000), ie those vehicles that are public service vehicles and require a certificate of initial fitness but are not used on local or scheduled services (eg touring coaches and community transport). It recommends that every wheelchair should be secured in the vehicle and it sets performance requirements for such equipment.
In addition, the Disabled Persons Transport Advisory Committee (DPTAC - the Government's statutory adviser on the transport needs of disabled people) produced The Recommended Specification for Buses Used to Operate Local Services in 1988, revised 1995. With the development of low floor vehicles, DPTAC produced a new bus specification in 1997 to complement the existing one, and to cover features required for fully accessible vehicles. It is the view of DPTAC that the PSVAR and the Department for Transport's associated guidance document supercede the above DPTAC specifications. However, the PSVAR does not apply to vehicles with fewer than 23 passengers and therefore in December 2001 DPTAC issued their Accessibility Specification for Small Buses designed to carry 9 to 22 passengers (inclusive).
The 'Department for Transport's Agreed Requirements - guidance notes for Vehicle Examiners' is a development of recommendations by the Vehicle Inspectorate to assist their examiners when considering the requirements appropriate to a particular vehicle. The Agreed Requirements are currently limited to the carriage of unrestrained wheelchairs in vehicles that carry standing passengers and not issued with an Accessibility Certificate under PSVAR. For vehicles without standing passengers and not issued with an Accessibility Certificate, the requirements of VSE 87/1 are applied.
While it is not a regulation, a relevant piece of work was carried out under the 'COST' programme sponsored by the European Commission. The COST programme supports co-ordination of research activities between different organisations, but does not fund the research itself. COST 322 addressed the subject of Low Floor Buses with the key objective of gathering information on current European operational experience in order to draw up guidance on best practice. The report provides guidance for the vehicle, the infrastructure and training, but no recommendations relate specifically to safety in the event of an accident.
ECE Regulation 25 specifies requirements for the strength and energy absorbing qualities of head restraints in vehicles. ECE Regulation 17 contains the same requirements for head restraints but in the context of a document with a wider scope. Hence throughout the report reference will be made to ECE Regulation 17 on the understanding that the two regulations are equivalent in this respect.
The DfT wishes to ensure that an appropriate level of safety is afforded to wheelchair users when travelling on public transport and that their needs are appropriately considered in legislation where necessary. This report summarises the results of a research programme devised to identify their level of safety compared to other passengers seated in the vehicle, and to recommend, where necessary, changes in legislation to improve that safety.
The scope of the research project was to compare the level of safety afforded to people seated in their wheelchairs with that afforded to non-disabled people (referred to throughout as 'vehicle seated occupants') also sitting in the vehicle. By examining the level of safety intended to be afforded to vehicle seated passengers in the safety and crashworthiness directives and regulations, it is possible to make observations regarding the level of protection afforded to wheelchair users in vehicles. The project covered frontal impacts only to M1, M2 and M3 vehicles.
To fulfil the objectives of the project, two distinct but complimentary phases of work were carried out. The first phase, a simulation study, was conducted to identify the main factors that influence the safety of wheelchair users in an accident. This was achieved by performing parameter sweeps of the most influential factors such as the wheelchair type, the occupant size, and the wheelchair restraint type, for each vehicle category considered. In addition, the direction in which occupants face within the vehicles, i.e. rearward or forward facing, was also considered. This study provided the information necessary for determining which dynamic tests to perform during the second phase.
The second phase comprised a series of sled tests to compare the level of protection provided to wheelchair seated occupants in comparison with those occupants in vehicle seats. It is from this research that recommendations for vehicle legislation can be derived.
The project considered the safety of wheelchair occupants when travelling in M category vehicles which are defined according to the European directive 92/53/EEC. M category motor vehicles with at least four wheels used for the carriage of passengers, are categorised as follows:
|
M1 |
= 8 seats in addition to the driver's seat |
|
M2 |
> 8 seats in addition to the driver's seat and a maximum mass = 5 tonnes |
|
M3 |
> 8 seats in addition to the driver's seat and a maximum mass > 5 tonnes |
For each vehicle class a crash pulse was chosen to characterise the occupant compartment deceleration in the event of a severe accident. This choice was based on the crash pulses used in current legislative test procedures for vehicle seated occupants and involved close consultation with the DfT. The pulses for each vehicle category are given below:
M1 Vehicles
UN/ECE Regulation 44, Figure1 (as used for testing child restraint systems). This pulse was selected on the basis that the deceleration corridor is derived from full scale M1 vehicle crash tests. This was verified by comparing the chassis deceleration recorded from various MPV's in the EuroNCAP programme, carried out at TRL.
M2 Vehicles
UN/ECE R44 as for M1 vehicles.
Based on the results of an accident study, Lawrence (1996) determined that a delta V of 48 km/h addressed 50% of all accidents in which an occupant was fatally or seriously injured. The chassis deceleration during a full-scale monocoque minibus crash test at this velocity, carried out at TRL, was found to be similar to the deceleration corridor in ECE R44. Lawrence therefore recommends that the R44 sled pulse should be used as a basis for sled testing minibus (M2 vehicle) restraint systems. This pulse was taken as representative of all M2 category vehicles.
The test pulse required by ISO 10542 and also PSVAR for M2 vehicles falls within the above corridor in all but one respect: both documents require that 15g is maintained for a minimum of 40ms, which means that an R44 pulse towards the lower bound of the above corridor may not comply with PSVAR.
M3 Vehicles
UN/ECE Regulation 80 is used for approval of the seats and their anchorages in large passenger (M3) vehicles. It includes a dynamic test of the seats and Figure 2 shows the deceleration corridor given in the regulation. Being the conventional pulse used for this type of vehicle, it was again applied within this work. It should also be noted that the R80 pulse has been adopted in Directive 96/37 for seat strength.
Four types of wheelchair design were investigated in the project following advice from the Medical Devices Agency (MDA). The wheelchairs were chosen to reflect the range of masses, stiffnesses and shapes of typical designs. The four wheelchairs selected for testing were:
Throughout the remainder of the report these wheelchairs will be referred to as the manual, electric, heavy electric and surrogate wheelchairs, respectively. It was understood that these wheelchairs represented a limited cross section of the available wheelchair designs and that additional design features unaccounted for in these four wheelchairs could affect the results and ultimately the conclusions derived from the investigations. However, it was beyond the scope of the current project to investigate additional wheelchair designs and the proposed selection of chairs was thought to cover the widest range of wheelchair features considered to be of greatest importance in vehicle impacts (i.e. wheelchair mass and stiffness).
The manual wheelchair was a production model, as shown in Figure 3. This is a standard folding wheelchair with a sling canvas seat, weighing 18 kg. The electric wheelchair was a production model weighing 57 kg as shown in Figure 4. The heavy electric wheelchair was a production model weighing 120 kg, shown in Figure 5. The surrogate wheelchair is described in ISO 10542 part 1 and Appendix D. This wheelchair is regularly used by TRL for dynamic testing of wheelchair restraint systems and is shown in Figure 6.
From the Hybrid III family of dummies, the 5th percentile female, the 50th percentile male and the 95th percentile male were selected. These dummies were used to represent the wheelchair occupant for both the modelling and testing work. The reason for this choice was that the Hybrid III family of dummies, and mainly the 50th percentile male, are used for most frontal impact legislative crash testing and hence can be regarded as the current world standard for this type of testing.
It was originally proposed that the BIORID dummy would be used to assess occupant safety in all rear facing impacts as it is more biofidelic than the Hybrid III dummy, having both a flexible spine and a biofidelic flexible neck. However, the BIORID dummy has been specifically designed and tested for impact speeds lower than 25 km/h, whereas the R44 and R80 pulses chosen for the investigations represent impact speeds of 50 km/h and 30 km/h, respectively. Consequently, there were concerns that the BIORID dummy could be damaged if used in these tests and that there would also be difficulties with the interpretation of test results given that the dummy has not been validated at these high impact speeds.
A suggested compromise was to use a Hybrid III dummy with a T-RID neck. However, as with the BIORID dummy, this dummy configuration was designed and validated at medium to low impact speeds. Discussions with the neck developers at TNO established that although the T-RID neck is more biofidelic than the Hybrid III dummy neck at low to medium severity impacts, it would be expected that the response of a restrained T-RID neck will match that of the Hybrid III neck at high impacts. Unrestrained, it is thought that the response of the Hybrid III neck would be more biofidelic than the T-RID neck. As a result of these discussions it was decided that, although not ideal, the complete Hybrid III dummy would be used for the rear facing impact investigations for both the modelling and the testing phases of the project. However, it must be remembered that the HYBRIDIII dummy was developed for forward facing frontal impacts and as such its biofidelity for rear impacts cannot be taken for granted. Hence, any injury criteria values measured for rear impact should be treated with caution as they may not be representative of real injuries.
The tests were carried out using Hybrid III 5th, 50th and 95th percentile dummies. The dummies were fitted with triaxial accelerometers in the head, chest and pelvis and fore/aft accelerometers on the upper and lower neck. The dummies also contained upper neck load cells and a chest potentiometer, and the 50th percentile was fitted with a lumbar spine load cell. In tests where the dummy was restrained by an occupant restraint the belt loads were also recorded. The masses of the dummies are given in Table 1.
Table 1: Hybrid III Dummy Masses
|
Dummy |
Mass (kg) |
|---|---|
|
5th |
50 |
|
50th |
75 |
|
95th |
100 |
A number of injury criteria with associated performance limit values were used to assess the likelihood of the occurrence of injury. These values were used to interpret the results obtained from both the modelling and testing studies. Where available, the criteria and performance limit values for the 50th percentile male were taken from current legislation such as the Directive on frontal impact, 96/79/EC. In order to offer equivalent levels of protection in accidents for wheelchair seated occupants compared with vehicle seated occupants, future vehicle safety legislation will need to specify performance limits that take into consideration the potential loadings on the vehicle from wheelchairs and their restraint systems. Additional criteria, especially for the 5th and 95th percentile dummies, were obtained from the literature. A summary of the injury criteria and associated performance limit values is given in Appendix 1.
The Neck Injury Criteria (NIC) was considered for assessing the potential for whiplash injuries in the neck under rear impact. However, because of the high severity of the impacts used in this project, it was found that NIC would not be suitable as it is generally only applicable for low to medium severity rear impacts. In addition, no current legislation uses the Neck Injury Criterion. In the absence of injury criteria for high speed rear impacts and because this was a comparative study, the injury criteria for the Hybrid III under frontal impact were used.
The Dynamic Restraint Test Facility (DRTF), at TRL, was used for the test programme. The DRTF comprises a rail mounted sled which is accelerated by elastic cords and decelerated by polyurethane deceleration tubes and olives. Dummy signal data was recorded by a Kayser-Threde data acquisition system. The data were analysed to ISO 6487 (2000).
Kinematic motion of both dummy and sled set up throughout the event were recorded using high speed video equipment (1000 fps) and two high speed cine cameras (400 fps). One camera showed the lateral view of the dummy at the point of impact and during its subsequent motion. The additional camera, where possible, showed the longitudinal view.
All of the computer models used for the project were developed and run under MADYMO version 5.4.1. MADYMO is a proprietary software package which analyses the dynamic response of moving systems through the idealisation of the structure into rigid and flexible bodies connected by joints. Surfaces can be attached to these bodies which can then be used to show how the bodies interact with each other. MADYMO also has the capacity to utilise Finite Elements (FE), whereby the structure is divided into many shapes that are subject to certain conditions or limitations in order to represent the surface structure of an object more accurately. The programme uses the equations of motion to calculate the dynamic interaction and the forces involved. MADYMO is recognised internationally as 'state of the art' and is widely used in the automotive industry for the simulation of occupant kinematics.
For all the models, contact stiffnesses were defined between:
Where experimental data was lacking, the contact stiffnesses were either estimated or derived from other MADYMO models.
TRL has previously conducted a field study of longitudinal and lateral accelerations on buses. Accelerations were measured on a range of vehicles over routes known to be particularly difficult to negotiate. Lateral accelerations over 0.2g were recorded on all the journeys in the study and in 18% of cases, accelerations over 0.35g were measured. In a few instances the acceleration exceeded 0.4g.
In general, the larger buses sustained fewer lateral accelerations over 0.35g, and low floor buses were marginally better than conventional buses.
A study was undertaken as part of this project to determine the extent of wheelchair movement on board a large bus (M3) under normal driving conditions, i.e. non-impact conditions.
An 11.5 m long Optare Excel, which is compliant with the Public Service Vehicles Accessibility Regulations 2000, was used for these tests. The regulations allow the wheelchair user to travel unrestrained in a rear facing position against a head and back restraint.
The vehicle was driven in a semicircle with a radius of 20 metres, at a velocity of 38 - 40 kilometres per hour.
The wheelchair used for these experiments was the PSVAR reference chair.
Twenty-eight analyses of the numerical model were carried out as described in Appendix 2. Various combinations of wheelchair, dummy size and restraint system were examined along with an investigation into the influence of the deceleration level within the ECE R44 corridor.
The results suggested that a floor anchored occupant restraint might be less favourable, compared with an upper anchorage location. This was based on both the dummy kinematics and the predicted occupant loading during the impact. When using the floor mounted restraint, the upper body rotated forwards about the waist, resulting in a high head excursion and in some cases head contact with the legs. In general, the dummy forces and accelerations were higher also, and there were more instances where the injury limits were exceeded.
There was not a significant difference in the dummy load levels between different wheelchairs. The method of wheelchair restraint was also found to have little effect, although greater wheelchair excursion was predicted when clamps were used.
The simulations predicted that dummy excursion and loading were linked to occupant size. In general, analyses with the Hybrid III 95th percentile predicted the highest values of these parameters, but the injury limits for this dummy were also greater and so the higher readings did not necessarily indicate greater injury risk.
Finally, equivalent models were subjected to two different deceleration pulses, both within the ECE R44 corridor. The results indicated that although the R44 corridor allows for acceleration variations of up to 8g, differences of only 5g in the peak deceleration level could noticeably affect the predicted dummy loads. This finding should be borne in mind when interpreting the predictions.
To further investigate forward facing occupants in M1 and M2 category vehicles a series of dynamic tests were carried out. The findings from the simulation work were considered when planning the test programme.
The primary objective for the test series was to assess whether the wheelchair seated occupant was provided with an equivalent level of safety as the vehicle seated occupant, through the use of instrumented dummies to compare the loading on the occupant.
In addition to this, the effect of diagonal belt anchor location was examined. The simulation work suggested that use of a floor anchored occupant restraint could have a negative effect on occupant protection when compared with an upper anchorage location. It was therefore necessary to revisit this issue in the test series.
The effects of a head and back restraint were also investigated to determine whether this could improve the protection for the occupant with respect to neck extension and movement in rebound.
The occupant space required within the vehicle was also assessed along with the loads that the vehicle anchorages would be required to withstand, in order to set requirements for vehicles.
In order to investigate M1 and M2 vehicle requirements eight sled tests with various set ups were carried out.
Six of the eight tests investigated occupant loading. A vehicle seated baseline test was completed for comparative purposes using a Hybrid III 50th percentile dummy in a commercially available seat that incorporated a three-point occupant restraint and head restraint integrated into the seat. This is shown in Figure 7. Figure 8 shows a typical set up for the wheelchair seated occupant.
Two tests investigated the loadings to the vehicle anchorage systems. Two different occupant restraint anchorage locations were used, one floor and one upper, and the wheelchair was restrained with a four point tie-down. In order to create the worse case situation for the vehicle anchorages the heavy electric wheelchair was used with the 95th percentile dummy. These set ups are shown in Figure 9 and Figure 10.
The vehicle environment for a minibus and taxi was represented on the sled. A production model minibus seat was used for the vehicle seated occupant test which incorporated an integrated 3-point restraint system.
The wheelchairs were restrained by a four point webbing tie-down secured to the floor by purpose designed Aluminium Track Fittings (ATF). The dummy was restrained independently by a lap and diagonal inertia restraint in all tests. Two types of wheelchair occupant restraint were investigated to compare the effects of different shoulder belt anchor locations. These were anchored to either an upper location or the floor.
The test set up for measurement of the restraint anchorage loading required additional instrumentation adjacent to all anchor locations in order to record the forces generated during the impact. Markers were positioned to enable measurement of the restraint angles at peak loading.
All wheelchair tie-down and occupant restraints were installed according to the manufacturer's instructions and the ISO standards.
Table 2 details the various test configurations used for the forward facing M1 and M2 vehicle research.
Table 2: Test Matrix - M1 and M2 forward facing
|
Test |
Dummy Seating Position |
Wheelchair Tie-down |
Dummy |
Occupant Restraint Diagonal Belt Anchorage Location |
Head/Back Restraint |
|---|---|---|---|---|---|
|
1 |
Minibus Seat |
N/A |
Hybrid III 50th |
Integral 3 point |
N/A |
|
2 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Floor |
No |
|
3 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Upper |
No |
|
4* |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Upper |
Yes |
|
5 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Upper |
Yes |
|
6 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Upper |
With Gap |
|
7 |
Heavy Electric Wheelchair |
4 Point Webbing |
Hybrid III 95th |
Upper |
No |
|
8 |
Heavy Electric Wheelchair |
4 Point Webbing |
Hybrid III 95th |
Floor |
No |
Twelve analyses of the numerical model were carried out for this condition as described in detail in Appendix 2. The model included a representation of a typical vehicle environment around the wheelchair based on a purpose built taxi. The wheelchair was positioned rear facing against the bulkhead separating the driver and passenger compartments.
Manual, electric and surrogate wheelchair models were compared as part of the study and additional analyses were carried out with a head and back restraint to determine whether this would reduce occupant loading.
The results suggested that a head and back restraint would be needed if the injury criteria limits were not to be exceeded as there was excessive rearwards head movement and over extension of the neck predicted by most of the analyses. The recommendations for the testing programme were to compare the protection afforded to the wheelchair occupant with and without a head and back restraint, with consideration being given to the energy absorption characteristics of the head and back restraint.
To further investigate rear facing occupants in M1 and M2 vehicles a series of dynamic tests were carried out. The findings from the simulation work were considered when planning the test programme.
The primary objective in the test series was to assess whether the wheelchair seated occupant was provided with the same level of safety as the vehicle seated occupant, through the use of instrumented dummies to compare the occupant loading.
In addition, the effects of occupant size and wheelchair stiffness were investigated. The use of a head and back restraint was also examined, to see whether its use could improve the protection provided for the occupant with respect to neck extension and movement.
Finally, a test was carried out to determine the dynamic strength requirements for a head and back restraint for rear facing wheelchair users in M1 and M2 vehicles.
In order to investigate rear facing M1 and M2 vehicle requirements, ten sled tests with various configurations were carried out (see test matrix, Table 5).
Tests 1 - 9 concerned rear facing occupant loading. A vehicle seated baseline test was carried out for comparative purposes using a Hybrid III 50th percentile dummy in a fold down seat and restrained with a three point belt. This is shown in Figure 15. Figure 16 shows a typical set up for the wheelchair seated occupant.
Test 10 examined the loading on a head and back restraint in order to define strength requirements for rear facing head and back restraints in M1 and M2 vehicles. A 50th percentile Hybrid II dummy was seated in a heavy electric wheelchair. The wheelchair was positioned rear facing against three force measuring plates that corresponded to the dummy head, torso and base of the wheelchair. The set up for the test is shown in Figure 17.
For the vehicle and wheelchair seated comparisons, the investigations into occupant size and wheelchair stiffness and the initial work on head and back restraint, a taxi bodyshell was mounted on the sled. For the remaining tests, the wheelchair and restraint system only were mounted on the sled.
The wheelchairs were restrained by a 2 point webbing tie-down system. This consisted of a Y-shaped heavy duty webbing strap that attached to the rear of the wheelchair by means of two large hooks. The dummy was restrained independently with a three point lap and diagonal inertia restraint.
In order to compare the safety of wheelchair seated occupants with the safety of vehicle seated occupants the occupant loadings have been expressed as a percentage of the injury threshold values. Where there are no injury criteria limits the vehicle seated results have been used as a base line (i.e. 100%), these results are shown in Table 6.
Table 6: Injury criteria comparison
|
50%ile injury limit |
Vehicle Seated |
Manual Wheelchair |
Surrogate Wheelchair |
|
|---|---|---|---|---|
|
Head Acceleration |
80g |
97% |
121% |
146% |
|
Neck Tension |
Vehicle Seated Baseline |
100% |
199% |
179% |
|
Lumbar Spine Compression |
Vehicle Seated Baseline |
100% |
141% |
207% |
|
Chest Acceleration |
60g |
55% |
66% |
48% |
In general, the wheelchair seated occupant was at greater risk of injury than the vehicle seated occupant. All injury criteria showed an increased level of risk up to double that of an occupant seated in a baseline vehicle seat fitted with a head restraint.
The results in the table indicate that the dummy head accelerations were greater for the wheelchair seated occupants and exceeded the injury criteria limits by 21% for the manual wheelchair and 46% for the surrogate. The corresponding vehicle seated result was just within the injury limit. The greater head acceleration for the surrogate wheelchair occupant was a result of head contact with the roof.
The level of tensile loading in the neck was significantly greater for the wheelchair seated occupants. However, the dummy kinematics indicate that neck injury is a concern for all rear facing occupants. When seated in the vehicle fold down seat without a head restraint, the dummy's head shattered the glass partition that separated the driver and passenger compartments, resulting in extension of the neck.
The injury criterion for chest acceleration was not exceeded. However the manual wheelchair seated occupant received greater chest accelerations than the vehicle seated occupant. This was due to the dummy loading the bulkhead after the stitching in the canvas seat back of the manual wheelchair partially failed.
The level of lumbar spine compression was greatest for the wheelchair seated occupants. However no injury criteria exist for the Hybrid III lumbar spine, so it is not possible to determine whether this would lead to lumbar spine injury.
The comparison of rear facing vehicle and wheelchair seated occupants indicates that head acceleration and neck loading are the primary areas of concern for the wheelchair seated occupants.
This was not examined for M1 and M2 rear facing.
Table 7 shows the important results from the tests examining the effect of occupant size. This was investigated using the 5th percentile small female dummy seated in the surrogate wheelchair. The loadings have been expressed as a percentage of the injury criteria.
Table 7: Occupant size comparison
|
Target Limit |
Results |
|||
|---|---|---|---|---|
|
5th % |
50th % |
5th % |
50th % |
|
|
Head Acceleration |
[80g] |
80g |
167% |
146% |
|
HIC |
1113 |
1000 |
165% |
105% |
|
Chest Acceleration |
73g |
60g |
123% |
48% |
Ten model analyses of the numerical model were completed and are described in detail in Appendix 2. Various combinations of wheelchair, dummy and restraints were investigated.
The simulation work suggested that there would be a low likelihood of serious injury in this type of impact as the dummy loads were well below the injury thresholds in most cases. The wheelchair tie-down system had little effect on dummy loading when the diagonal part of the belt was mounted in the upper location. The effect of using floor mounted occupant restraints was not investigated in the simulation work.
To further investigate forward facing occupants in M3 category vehicles a series of dynamic tests were carried out. The findings from the simulation work were considered when planning the test programme.
The primary objective for forward facing occupants in M3 category vehicles was to assess whether the wheelchair seated occupant was provided with an equivalent level of safety as the vehicle seated occupant, through the use of instrumented dummies to compare the occupant loading.
In the dynamic testing, along with a comparison of the safety of vehicle and wheelchair seated passengers the different combinations of wheelchair restraint and occupant restraint geometry were investigated. This was examined to determine whether there was a negative effect on the protection to the occupant. Wheelchair stiffness and occupant size were also investigated.
The effect of a head and back restraint was then investigated to determine whether this could improve the protection provided for the occupant with respect to neck extension and movement in rebound.
In order to set requirements for vehicles, the occupant space required within the vehicle was examined along with the loads that the vehicle anchorages would have to withstand.
In order to investigate M3 vehicle requirements, thirteen sled tests were carried out with various configurations (see test matrix, Table 10).
Tests 1 - 11 concerned occupant loading in M3 vehicles. A vehicle seated baseline test was completed for comparative purposes using a Hybrid III 50th percentile dummy in a commercially available coach seat that included an integrated three-point occupant restraint and a head restraint. This is shown in Figure 21. A series of wheelchair seated tests then followed to investigate the issues outlined in the scope. Figure 22 shows a typical set up for the wheelchair seated occupant.
Table 10: Test Matrix - M3 forward facing
|
Test |
Dummy Seating Position |
Wheelchair Restraint |
Dummy |
Occupant Restraint Diagonal Belt Anchor Location |
Head and Back Restraint |
|---|---|---|---|---|---|
|
1 |
Coach Seat |
N/A |
Hybrid III 50th |
Integral 3 point |
No |
|
2 |
Surrogate Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Floor |
No |
|
3 |
Surrogate Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Upper |
No |
|
4 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Upper |
No |
|
5 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Floor |
No |
|
6 |
Manual Wheelchair |
Clamps |
Hybrid III 50th |
Upper |
No |
|
7 |
Manual Wheelchair |
Clamps |
Hybrid III 50th |
Floor |
No |
|
8 |
Surrogate Wheelchair |
4 Point Webbing |
Hybrid III 95th |
Upper |
No |
|
9 |
Surrogate Wheelchair |
4 Point Webbing |
Hybrid III 5th |
Upper |
No |
|
10 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Upper |
Yes |
|
11 |
Manual Wheelchair |
4 Point Webbing |
Hybrid III 50th |
Floor |
Yes with gap |
|
12 |
Heavy Electric Wheelchair |
4 Point Webbing |
Hybrid III 95th |
Upper |
No |
|
13 |
Heavy Electric Wheelchair |
4 Point Webbing |
Hybrid III 95th |
Floor |
No |
Six analyses of the numerical model were carried out and these are described in detail in Appendix 2.
The simulation work suggested that the occupant could experience high levels of neck bending and chest acceleration if seated in the surrogate wheelchair. The reason for this observation was that the stiff wheelchair structure keeps the dummy back away from the back restraint, hence allowing more relative rearward motion between the head and shoulders.
To further investigate rear facing occupants in M3 category vehicles a series of dynamic tests were carried out. The findings from the simulation work were considered when planning the test programme.
The primary objective for rear facing occupants in M3 vehicles was to assess whether the wheelchair seated occupant was provided with the same level of protection as the vehicle seated occupant, through the use of instrumented dummies to compare the loading to the occupant.
In order to investigate M3 vehicle requirements six sled tests were carried out with various configurations as shown in the test matrix, Table 15.
Occupant loading was investigated in tests 1 - 5. A vehicle seated baseline test was completed for comparative purposes using an unrestrained Hybrid III 50th percentile dummy in a commercially available rear facing fold down seat as shown in Figure 29. Figure 30 shows a typical set up for the wheelchair seated occupant in which the wheelchair and dummy are both unrestrained.
Test 6 examined the loading on a head and back restraint in order to define strength requirements for rear facing head and back restraints in M3 category vehicles. A 50th percentile Hybrid II dummy was seated in a heavy electric wheelchair and the wheelchair was positioned rear facing against three force measuring plates that corresponded to the dummy head, torso and base of the wheelchair. The set up is shown in Figure 31.
Various M3 vehicle environments were created on the sled.
Table 12: Test Matrix - M3 rear facing
|
Test |
Dummy Seating Position |
Wheelchair Restraint |
Dummy |
Occupant Restraint Diagonal Belt Anchor Location |
Head and Back Restraint |
|---|---|---|---|---|---|
|
1 |
Fold down seat |
N/A |
Hybrid III 50th |
none |
Yes |
|
2 |
Surrogate Wheelchair |
none |
Hybrid III 50th |
none |
Yes |
|
3 |
Manual Wheelchair |
none |
Hybrid III 50th |
none |
Yes |
|
4 |
Manual Wheelchair |
2 Point Webbing |
Hybrid III 50th |
Upper |
Yes |
|
5 |
Manual Wheelchair |
2 Point Webbing |
Hybrid III 50th |
Upper |
Yes with gap |
|
6 |
Heavy Electric Wheelchair |
2 Point Webbing |
Hybrid II 50th |
Upper |
Yes |
The Public Service Vehicles Accessibility Regulations 2000 allow a wheelchair user in a bus to travel unrestrained, in a rear facing position, against a back restraint or bulkhead. It is assumed that this configuration will adequately keep the wheelchair in the designated space during normal transit, and this is how many thousands of journeys take place without incident.
The regulations demand a method for restricting lateral movement of the wheelchair into the gangway. This can be a vertical stanchion situated at the front end of the wheelchair space and running continuously from the floor to the roof, or a retractable rail extending from the front of the wheelchair space. A range for the position of both these items within the wheelchair space is specified.
A series of trials were carried out to study the movement of an occupied wheelchair in normal transit on a bus that is compliant with the PSVAR.
Previous research at TRL has demonstrated that lateral accelerations on low floor buses can reach 0.4g on bus routes selected as being difficult to negotiate. With this knowledge, an Optare Excel 11.5 was tested on the large central area of the TRL research track with a dummy seated in a wheelchair, to determine whether the wheelchair was displaced during manoeuvres that generated this level of lateral acceleration.
The Optare Excel is compliant with the PSVAR and is fitted with a cranked vertical stanchion (see Figure 34). This feature is to allow easier access into the wheelchair space and provides a more convenient hand hold position when compared with a purely vertical stanchion. It was necessary to test other possible systems, so the stanchion was also removed from the vehicle, and a rail fitted in the required position. It was not possible to obtain a production rail that was compliant with the PSVAR within the time scale of the project, so a mock up was used in its place.
To generate the level of lateral acceleration required for the experiment the vehicle was driven at a constant velocity of 24-25 mph. A left turn was then executed, of 20 metres constant radius, while the speed was maintained. This gave a repeatable test procedure suitable for the experiment. The acceleration at the wheelchair space was logged and downloaded for analysis.
The experiment was carried out with a Hybrid II 50th percentile dummy seated in a DDA reference wheelchair. The dummy and wheelchair were rear facing and unrestrained, with the brakes applied. The seat back of the wheelchair was in contact with the back restraint.
The stanchion was tested in three positions (see Figure 35). These were 400, 480 and 560 mm from the front end of the wheelchair space, measured at the base of the stanchion. Due to the difficulty in maintaining a constant speed whilst conducting the prescribed turn, each position was tested a number of times to ensure that the driver had been able to complete the manoeuvre successfully, with the correct level of lateral acceleration. In all, 12 test runs were completed.
For the retractable rail, a section of tubing of 34mm diameter was attached to the luggage pen of the vehicle, for convenience, and extended to a point 540mm from the front of the wheelchair space (see Figure 36). To comply with the PSVAR, the rail was initially fitted between 600 - 800 mm from the floor of the vehicle, and moved in intervals of 50 mm.
A VHS camera was fitted on board. This was used to record the wheelchair movement during the test and to assess the extent to which the wheelchair moves into the gangway. If the wheelchair was maintained within the wheelchair space and the dummy remained in the wheelchair during a turn that registered 0.4g on the logging equipment, then the stanchion or rail could be said to be performing satisfactorily.
Both the stanchion and rail are designed to prevent movement of the wheelchair into the gangway, where it could become a hazard for the wheelchair passenger and standing passengers in the vicinity. This is necessary in the Optare Excel and similar vehicles, because as the bus turns left there is a weight transfer to the left of the wheelchair, adjacent to the gangway. This reduces the grip of the wheels on the right hand side of the wheelchair, beside the wall. The mass can then pivot about the vertical axis of the left rear wheel. This effect is exaggerated by the front wheels, which are on castors, and without brakes.
The film of the tests indicates that the motion described could occur when lateral acceleration reached 0.29 - 0.3g, although movement into the gangway did not occur unless levels of acceleration reached 0.33 - 0.34g. It is only possible to give an approximation of when the motion occurred as the accelerometer and video were not directly linked.
The Public Service Vehicles Accessibility Regulations 2000 state that the base of the stanchion must be between 400 and 560mm from the front of the wheelchair space. During the tests with the stanchion at 400mm almost no wheelchair movement was observed, even when the lateral acceleration recorded in the vehicle reached 0.4g. The same was true when the stanchion was moved forward to 480mm. The wheelchair could not move, because there was only a short distance of approximately 50mm laterally, between the handrim on the left rear wheel and the stanchion. Therefore as the wheelchair begins to rotate about the vertical axis of the wheel, the handrim contacted the stanchion and the movement ceased.
When the stanchion was positioned at 560mm from the front of the wheelchair space greater movement occurred as the stanchion was ahead of the leading edge of the handrim. However the stanchion contacted the front edge of the wheel and tyre and further movement was prevented. At no time did the wheelchair cross the plane of the wheelchair space and move into the gangway. As the stanchion was seen to perform satisfactorily in these tests, whilst positioned at the extremes of the range allowable, it was not necessary to conduct further tests outside this range.
A retractable rail is permitted in the PSVAR in the place of a stanchion. This must extend at least 540mm from the front of the wheelchair space and be at a height between 600 and 800mm from the floor. A mock up device was fitted in the bus and braced laterally so that it was capable of bearing the load required. The rail was initially positioned within the height range described above.
At 600mm the rail did not perform well in separate tests where the peak lateral acceleration reached 0.35 and 0.4g. The wheelchair rotated about the vertical axis of the left rear wheel but the rail did not restrain the wheelchair due to the gap in the armrest, see Figure 37. The left front wheel and foot plate partially intruded into the gangway. Further movement was only prevented when the rail contacted the dummy leg.
The test was repeated with the rail at a height of 650mm. At this position the wheelchair did not move when the acceleration was 0.4g, because the rail was at the same height as the padded armrest on the wheelchair. When the wheelchair began to move it quickly contacted the rail, which prevented any movement into the gangway. The rail was then raised to 700mm. This height exceeded any side structure on the wheelchair and during the test there was significant wheelchair motion into the gangway. The wheelchair rotated about the vertical axis of the left rear wheel and passed under the rail until it was arrested by contact with the forearm of the dummy, Figure 38. At this point, part of both the left front wheel and foot plate were outside the designated wheelchair space and in the gangway. The rail was then raised to 800mm from the floor, the maximum height allowed in the PSVAR. The wheelchair again rotated under the rail, this time resulting in the greatest excursion into the gangway of the vehicle. At the end of tests where peak lateral accelerations were in the range 0.33 - 0.4g, the entire left foot plate and the left front wheel were outside the wheelchair space.
The tests described above were carried out with the rail at a height that was compliant with the PSVAR. It was apparent during these experiments that the rail was not performing satisfactorily at most of the heights tested, in the range allowed. At 700mm or above the reference wheelchair could pass underneath the rail, therefore there was no benefit in conducting further tests above the 800mm upper limit. Instead, additional tests were carried out with the rail below the lower limit, at 550mm from the floor. During these tests the rail performed well, and prevented the wheelchair from rotating into the gangway. The rail was more effective at this height because as movement began the rail contacted the handrims and restrained the wheelchair from further movement.
The results have shown that the reference wheelchair will move when exposed to levels of lateral vehicle acceleration that are possible on bus routes. Furthermore the effectiveness of the available methods for restricting this movement is variable, and dependent on the location of the device.
During tests with a vertical stanchion, wheelchair movement was restricted in all tests, when the position of the stanchion was varied within the range allowed. The level of peak lateral acceleration exceeded 0.4g in some of these tests, but the wheelchair did not move into the gangway. The optimum position for a stanchion is ahead of the vertical axis of the rear wheel as the wheelchair tends to pivot around this axis at levels of acceleration that were approaching 0.35g. However it should not be forward of the leading edge of the rear wheel because it is desirable for the wheel handrim and stanchion to make contact.
The performance of the rail in these tests was of concern. The ability of the device to restrict wheelchair movement was highly dependent on its height. If the rail could interact with the side of the wheelchair after movement began then it could restrict the motion sufficiently to keep the wheelchair and dummy out of the gangway. However, when tests were carried out at different heights allowed by the PSVAR, the wheelchair was able to rotate underneath the rail when subjected to lateral vehicle accelerations of the magnitude possible on bus routes. Based on these findings, the use of a single bar horizontal rail is not recommended, although a design with depth or some form of adjustment may be acceptable.
The results can only be used to give an indication of the performance of the stanchion or rail with respect to the DDA reference wheelchair only. Caution should be exercised when drawing conclusions for other wheelchair types. For instance, the stanchion performed well due to its position in relation to the large rear wheel, enabling it to stop the pivoting action of the wheelchair. The results may have been different for an attendant controlled wheelchair or an electric wheelchair with smaller rear wheels. The ability of the rail to restrict wheelchair movement was highly dependent on its height from the floor of the bus. It is therefore difficult to make recommendations regarding the effectiveness of the device using a single wheelchair type.
The only movement observed was a pivoting about the vertical axis of the left rear wheel; at no time did the wheelchair move forward. The brake performance could have been a factor in the lack of forward excursion of the wheelchair. The brake mechanism of the reference wheelchair appears to be more robust than that of a standard manual wheelchair. In these tests, tyre pressure was set at the lower limit of the recommended inflation level. The action of the brake shoe pressing on the tyre makes correct inflation pressure important. In the real world wheelchair tyres may not be fully inflated. Furthermore, there are occasions when a vehicle is likely to be accelerating whilst executing a turn, for instance entering a roundabout. It was not appropriate to investigate the possible effects of vehicle acceleration during the manoeuvre, as this could not be done with sufficient repeatability. However, it is possible that there would be a weight transfer to the front of the wheelchair, reducing the grip of the rear wheels.
In summary, the vertical stanchion appeared to provide a better means of restraining the wheelchair, although this conclusion requires validation against other wheelchair types, especially those with smaller wheels. The performance of the moveable horizontal rail was very dependent upon the dimensions of the wheelchair involved, and it is hard to see how the situation could be improved. This type of restraint therefore gives rise for concern, although again, tests with other designs of wheelchairs should be carried out before any actions are decided upon.
It is estimated that there are around 1,200,000 wheelchair users in this country representing approximately 2% of the population. If wheelchair users have the same social need to travel and the same ability to do so then it would be reasonable to assume that in all modes of transport wheelchairs users may represent 2% of all travellers. There is insufficient information on which to compare the travel patterns of wheelchair users with that of the population as a whole, nor is it possible to identify the proportion of wheelchair users who may transfer to a vehicle seat when travelling. However, it is unlikely that wheelchair users as group travel to the same extent as the population as a whole and we can be certain that not all will travel seated in a wheelchair. Furthermore, the travel patterns of wheelchair users may change as more vehicles become wheelchair accessible. In the absence of any information to the contrary it is assumed that wheelchair users who travel seated in their wheelchair will be exposed to 50% of the transport risks compared to other road users (i.e. 1% of the population). On this basis the overall exposure to travel risks may be taken as being proportionate to that of other road users for any type of vehicle that may be wheelchair accessible.
Cars, taxis and private hire vehicles with no more than 8 passenger seats are all M1 category vehicles 1 . Accident data does not differentiate between these vehicles and therefore they must be considered as a whole. As such they are all referred to as cars. Whilst there may be only a small proportion of private hire vehicles that are wheelchair accessible, many taxis are wheelchair accessible and there is a growing supply of wheelchair accessible cars for private use, for patient transport and similar applications.
For the year 2000 RAGB accident data shows 1,665 car drivers and passengers killed and 18,054 seriously injured. This is a slight improvement on both the 94-98 baseline average of 1,762 and 21,492, and the 96-2000 average of 1,730 and 20,071 respectively. If, in future, 1% of these are wheelchair users there is the potential for them to be involved in 180 serious injuries and 16 fatalities per year based on the data for the year 2000. The findings of this research project indicate that improvements are needed if wheelchair users are to be afforded an equivalent level of protection to that of other vehicle occupants. This suggests that making no changes is likely to result in a higher number of wheelchair user casualties than is estimated here. However, there is no evidence to suggest that such levels of injury are occurring at present. This suggests that either current travel patterns are lower than they are estimated to be in the future or other factors, such as the nature of the journeys undertaken or the wheelchair location in a vehicle, have an influence on the exposure to transport risks.
M1 vehicles are already equipped with a means of securing a wheelchair and are fitted with a wheelchair occupant restraint. The significant additional factors recommended for future designs include the provision of a head and back restraint for rear facing occupants, technical requirements