Jeanne Breen, Executive Director, European Transport Safety Council

The safety of people walking in urban areas has now to be considered in many European

countries in the context of policies for encouraging people to travel on foot, by cycle or by

public transport rather than by car in order to reduce environmental damage, improve

public health, and enhance the quality of life in towns and cities.

The principal conclusion of a recent review by the European Transport Safety Council

(ETSC) was that by implementing known countermeasures it should be possible to achieve

considerable increases in the use of healthier and more environmentally friendly means of

transport and still reduce the numbers of deaths and injuries among pedestrians and cyclists.

However, further concerted action needs to be taken by policymakers at local, national and

international levels to ensure that this positive scenario can be brought about (ETSC,

1999a).

Practically everyone needs to walk whether for work, shopping, education or leisure and

making the pedestrian environment safer will affect many people. The core of this paper is

to set out the challenge that providing for safer walking presents to policymakers and

professionals concerned with the many relevant aspects of urban planning and design of

the road transport system and its use.

The aim is, also, to reflect the consensus which exists, at least between independent road

safety experts from across the European Union, about the strategies and priority measures

needed to reduce pedestrian crash injury risk against the background of policies to

increase levels of walking. Updates on developments in European Union and national

policies on pedestrian safety will be presented along the way.

The basis of this contribution are recent ETSC’s recent reviews on the Safety of

Pedestrians and Cyclists in Urban Areas and Priorities for EU Motor Vehicle Safety

Design, as well as the contributions on vulnerable road user safety made recently by the

OECD, ECMT, and the EU projects, MASTER, DUMAS, PROMISING and WALCYNG.

Wherever possible, the international statistical comparisons presented include New South

Wales and Australia in addition to EU countries. These indicate broadly similar

motorisation levels and per capita pedestrian death rates.

Survey data from a selection of seven European countries show that 15-30% of all trips

are made by walking, the highest figure being for Great Britain (PROMISING, 2001). For

short trips the share of walking can rise to 40%. This EU project also identified that:

decreasing since the early 1980s, which may be partly related both to the increased

travelling distances resulting from urban development, and to the increase in vehicle

ownership.

2

While car users comprise the greatest proportion of overall road deaths (57%), the risk of

death on EU roads is substantially higher for vulnerable road users. It has been estimated,

albeit roughly, that walking is around 9 times riskier than travel by car for the EU as a

whole. Pedestrian deaths comprise 15% of total road deaths, with the UK (25%) and the

Netherlands (10%) at either end of the range.

Motorcycle/moped 16

Foot 7.5

Cycle 6.3

Car 0.8

Bus and coach 0.08

Rail 0.04

Whether measured by rates of pedestrian deaths per 100,000 population or motor vehicles

which, of course, do not take any account of the level of the activity there are substantial

differences between Member States with the highest rates in Portugal and Greece and the

lowest rates in Sweden and the Netherlands. More information is needed about levels of

pedestrian and cyclist traffic in the EU, however, before crash risk differences can be fully

understood.

3

The average severity is generally higher in rural areas, but the great majority of casualties

to pedestrians occur in urban areas. The over-55 and under-12 age groups are those with

the highest risk of pedestrian injury. Risk from traffic consists mainly of risk from motor

vehicles – around 90% - and mainly from contact with the fronts of cars.

Accident analysis shows that about 50% of pedestrian deaths occur while crossing a road.

About a quarter occur while boarding or alighting from a bus or getting into or out of a car.

Others occur while walking along the road, playing, running, or working. Most fatal crashes

involving pedestrians are not located at a marked crossing, the vast majority occurring

more than 50m from such a crossing. Elderly people are most frequently hit by vehicles

when halfway or further across the street, while children are mostly hit when starting to

cross.

The overall long-term trend in deaths has been downward for pedestrians. Between 1980

and 1995, the pedestrian death rate per capita for the EU as a whole fell by 30%. Studies

4

indicate this may be due in some instances to a decline in walking (for example, amongst

children) as more people take to their cars for local journeys.

However, several Member States are now experiencing annual increases in pedestrian

deaths and encouragement is now being given in various countries to travel by foot,

bicycle or public transport. For example, the Danish National Traffic Plan states that 4% of

total car traffic should be converted into cycling and walking by the year 2005 and onethird

of all car traffic under 3 km into non-motorised travel. As travel by public transport is

also encouraged, increasing account needs to be taken of the safety of walking or cycling

to catch the bus, tram or train.

The ageing of the road user population experienced internationally (shown below) is also

likely to influence future trends and increase the need for action. For both Australia and

the European Union (15), over one fifth of the population will be 65 years or above by

2030. Despite the rising number of older driving licence holders in many countries,

declining driving ability and financial constraints mean that many motorists will have, at

some stage, to give up their car. A larger percentage of the older population will be

dependent on public transport which will involve pedestrian trips. The risk of death in EU

traffic for pedestrians aged 65 and older is currently four times higher than for young

adults.

AUSTRIA 15.4 17.8 20.1 25.2

BELGIUM 16.8 17.9 21 25.4

DENMARK 14.9 16.7 20.3 23

FINLAND 14.5 17.1 22.6 25.8

FRANCE 16 16.8 20.6 24

GERMANY 16.2 19.7 21.4 25.8

GREECE 17.3 19.5 21.8 25.4

IRELAND 11.3 12.2 15.3 18.7

ITALY 18.1 20.6 23.5 28.1

LUXEMBOURG 14 14.5 16.4 19.8

NETHERLANDS 13.6 15.3 19.7 23.9

PORTUGAL 15.4 16.9 19.4 22.8

SPAIN 16.9 18.4 21.2 26.4

SWEDEN 17.3 19.2 22.7 25.1

UNITED KINGDOM 15.7 16.7 19.6 23.5

USA 12.6 13.2 16.5 20

JAPAN 17 21.8 26.8 28.3

Source: US Bureau of the Census, International Databases February 2001 in Transport and Ageing Society, ECMT 2001

Most road safety problems for pedestrians are common to all European countries and

beyond. These result from a complex mix of factors. However, underlying all other

problems is the fact that the modern traffic system is designed largely from a car-user

perspective. Mass motorisation in much of Europe since the 1960s has created a traffic

system which caters mainly for motor vehicle users. Only since the 1980s has there been

understanding about the need for coherent planning of route networks for pedestrians and

5

only since the 1990s has long term planning for sustainable transport policies got off the

ground (OECD, 2001).

The following problems emerge as being key issues for improving pedestrian safety and

need to be addressed in combination in future traffic system planning.

vulnerable of citizens and are hit by the fronts of cars. Until safer car fronts are provided by

the car industry, the only protection available is clothing. Speed plays an important role in

determining the severity of the outcome of collisions. As Figure 6 illustrates, if the collision

speed exceeds 45 km/h the likelihood for a pedestrian to survive the crash is less than 50

per cent. If the collision speed is less than 30 km/h more than 90 per cent of those struck

survive. Speed management, therefore, is a key element in a safer traffic system for

vulnerable road users.

advantages of walking. In relation to other road users, however, this presents a problem. A

driver can never be sure when or where to expect a pedestrian.

and receive serious injuries just because of an uneven surface. The instability of

pedestrians is an even bigger problem when they are mixed with motor traffic.

be hidden by one. At night the problem is more severe. A parked car is the most commonly

cited source of obstruction.

reduced capability, and people with reduced mobility.

pedestrians it means extra muscular activity. They are, therefore, highly motivated to find

and keep to the easiest routes, often the most direct ones. Studies have shown that

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90

Teichgräber, 1983

Ashton, 1982

Waltz et al., 1983

6

pedestrians place a higher value on their time than drivers or those on board public

transport vehicles.

killed in a crash and tested for alcohol were found to have blood alcohol levels above the

legal limit for driving. This rate was higher than that of drivers involved in fatal crashes

(Fontaine et al., 1997).

the priority task, like window-shopping or chatting with friends. This, together with the fact

that the modern traffic environment is often designed for cars rather than for pedestrians,

creates a state of estrangement. Providing pedestrian facility is typically an afterthought

rather being considered as an integral part in the planning and design of the traffic system.

Even the majority of European car drivers believe that much consideration should be paid

to walking and cycling when planning for the future according to SARTRE, a survey of car

drivers conducted in 19 European countries in 1997.

In the development of EU and national targeted programmes, and most explicitly in those

embodying the sustainable safety or Vision Zero concepts, it is recognised increasingly

that preventing road death and disabling injury entails a traffic system that is better

adapted to the needs, errors and physical vulnerabilities of its users rather than one which

expects users to cope with increasingly demanding conditions. While challenging to

deliver, this approach is founded in pragmatism and ergonomics. Its innovation lies in

recognising that road death and severe public health loss is a feature of poor design; that it

can and should be avoided by putting to greater effect and implementing more widely in

targeted programmes, key safety principles and measures which have been known about

for many years.

The European Commission has accepted that such an approach is necessary to meet the

highly ambitious target which has just been set to reduce deaths by 50% by the year 2010

across the EU (CEC, 2001a). In their consultation on a new EU road safety programme

2002-2010, which is expected shortly, the Commission has already concluded that a better

balance is needed between the safety of vulnerable road users and the mobility of motor

vehicle users especially in urban and residential areas (CEC, 2001b).

In the next sections, the strategies and measures which European experts believe are key

to the delivery are set out and observations made on the efforts at EU, national or local

level. The focus is on evidence-based strategies and measures aimed at the provision of

safer environments through planning, infrastructure provision and vehicle design.

There are many ways in which transport policy in general and road safety policy in

particular can contribute to reducing crash injury risk from traffic for those travelling on foot.

ETSC has identified the key strategies for pedestrian safety as follows:

journeys

management framework

where conditions are favourable to separation.

7

area-wide speed and traffic management, increased pedestrian and vehicle

conspicuity, and vehicle engineering and technology.

training and the enforcement of traffic law.

encouraging them in steps that they can take to reduce their risk.

Land use planning can make a useful contribution to minimising pedestrian exposure to

risk of accident and injury. In planning the evolution of land-use, priority can be given to

locating the most likely destinations for walking and cycling - homes, schools, workplaces,

shops, social and recreational facilities, and public transport stops - where they can be

more readily served by safe, attractive and convenient routes for walking and cycling.

As sites and buildings are adapted, redeveloped or developed for the first time,

opportunities can be taken to achieve layouts which separate access by motor vehicles

from that on foot, and adapt the latter to the existing local network of pedestrian routes,

including routes from public transport stops.

Classifying the urban road network according to road function, setting appropriate speed

limits according to that road function and improving road layout and design to encourage

better use is now recognised, amongst EU Member States active in road safety, as

fundamental to urban safety management.

The Netherlands, in particular, has made considerable progress in establishing road

hierarchies and the UK and the Nordic countries have stated their intention to do this to

provide a better framework for area-wide risk reduction in their national road safety

strategies. The development of EU best practice guidelines on urban safety management,

amongst other themes, is foreseen in the new EU road safety programme, which is

expected to be announced this summer.

In the context of encouragement for walking, urban safety management needs to give high

priority first to identifying the pattern of journeys that people want to make on foot and then

to creating safe, attractive and connected routes for this pattern of journeys. These routes

should be designated in conjunction with the functions of each road for all kinds of road

user, and in particular so that motor traffic uses each road in ways that are consistent with

the safety and convenience of pedestrians.

Routes will typically consist of a mixture of sections of footpath separate from any

carriageway, wholly pedestrian areas with or without admission of cyclists, footways

alongside carriageways, and carriageways or other surfaces shared with motor vehicles.

Where routes cross appreciable flows of motor vehicles, careful attention will be given to

the location and design of the crossing point. Where the routes are not separated from

carriageways, and even more so where surfaces are shared with motor vehicles, the

layout will be such that the speeds of the latter are moderated.

Concentration of motor traffic onto main roads should enable the more local roads to be

adapted to enable them to perform their functions in respect of motor vehicles consistently

8

with their forming parts of safe and attractive routes for pedestrians and cyclists. The more

these roads are used for walking and cycling, the more aware drivers will become of the

likelihood of encountering pedestrians and cyclists, and thus the lower the risk that motor

vehicles will pose to them. On public transport routes, whether bus or light rail routes on

main or more local roads, or bus or rail services on segregated tracks, stopping places

should be served by the network of routes for walking and cycling.

A fresh look at road hierarchies in relation to pedestrian safety has been undertaken

recently by researchers coming together in the EU PROMISING project. It was based on

the requirements of coherence of the network, directness, safety, comfort and

attractiveness on the one hand and on the new concepts for road safety in the Dutch

sustainable traffic system and the Swedish Zero Vision on the other hand. The hierarchy

was developed only for built-up areas and is set out in Table 2.

With reference to this hierarchy, it is worth noting that the long debate about whether the

general speed limit to be favoured across Europe should be 50km/h or 60 km/h has been

largely resolved in favour of 50km/h, with increasing use of 30km/h off main roads. ETSC’s

comment on grade-separate crossings can be found in a later section.

Separation can take the form of pedestrian areas, footways alongside carriageways,

sections of footpath separate from the carriageway and grade-separated crossings.

Pedestrians need designated physical space with adequate pavement width such that

pedestrians need not walk on the carriageway and for those using wheelchairs.

Pedestrian areas may be designed as such or be conversions from streets used by

vehicles. Their value in improving safety has been demonstrated widely, especially for

shopping streets (e.g. DUMAS). Pedestrian areas may be exclusively for pedestrian use,

for pedestrians and cyclists or for pedestrians and cyclists along with some permitted

vehicles at certain times of the day. The facility for vehicles to use converted areas outside

times of closure will often remain for reasons of access and servicing, but the surface and

layout of the street are designed for pedestrians, with a clear indication of the paths to be

followed by vehicles when they have access.

While streets dominated by heavy flows of traffic tend to be threatening to pedestrians,

anxiety. Whilst the fear of personal crime may be out of proportion to its reality, this needs

to be considered in the layout and design of areas used by pedestrians, if they are to be

used.

By physically restricting access for vehicles, pedestrian zones create an environment

where travel on foot and by cycle is safer. Opinion on admission of cyclists to these areas

may be divided, but there is a need to avoid pedestrian areas resulting in unsafe or

inconvenient conditions for cyclists, for example by forcing them to use busy distributor

roads. In Mechelin cycling is permitted in pedestrian streets in order to avoid detours for

9

cyclists and evaluation had shown that this, so far, has proved to be safe (Dykstra et al,

1998). Research in the UK indicated that conflicts between cyclists and pedestrians in

pedestrian areas were less of a problem than appeared (Trevelyan and Morgan, 1993).

Segregating cyclists and pedestrians in pedestrian areas will not always be possible.

Where it is desirable, cycle movements can be combined with those of selected vehicles,

such as buses and service vehicles, permitted at particular times of day or channelled by

defined paths.

Pedestrians and cyclists are particularly at risk when crossing heavily trafficked roads and

are generally safer when separated from traffic. However, the benefits of grade-separated

crossings, which can be expensive in relative terms, are not always realised. To be

troublesome ramps and keep vulnerable road users on their natural desire-line whilst

motor vehicles undergo the changes in grade and level. The main use is for crossing roads

with speed limits of 60km/h or higher or heavily trafficked roads. Subways should be

brightly lit, regularly cleaned, have good through visibility and be consistently overlooked

(IHT, 1997).

Where separation can be achieved in ways which provide convenient and attractive routes

for all road users, it very largely removes risk from traffic in the areas of separation - but

this advantage may be offset by increased risk where road users re-enter shared space.

Integration of different kinds of road use by sharing of space often has the advantages of

requiring less adaptation of the roads and paths and enabling more direct routes to be

provided.

Taken together, the means of reducing risk require action to create safer conditions for

integrated use of shared road space through:

(a) managing speed and traffic through improving junction design and layout,

implementing area-wide treatments and speed zones and developing intelligent

speed adaptation and

(b) improving vehicle and user conspicuity.

Road safety engineering measures to create safer conditions for pedestrians can be

considered in terms of traffic reduction, speed reduction, junction treatments, the

redistribution of road space and the creation of special facilities.

trafficked routes for cyclists and a safer pedestrian environment as part of an area-wide

approach to avoid displaced traffic leading to more crashes elsewhere. Even at low

speeds, mixing with heavy traffic, especially lorries, is hazardous. The diversion of through

and unnecessary traffic from some areas will reduce potential conflict but will require

appropriate advance signing and, possibly, some road construction.

Speed of motor vehicles is critical to the safety of vulnerable road users. At low speeds

drivers have more time to react to the unexpected and avoid collisions. At speeds of below

30 km/h pedestrians can mix with motor vehicles in relative safety.

10

The development of speed management and traffic calming to deal with inappropriate

speed in urban areas in Europe has been documented by Kjemtrup and Herrstedt (1992).

These techniques comprise traffic management measures ranging from discouraging

traffic from entering certain areas to installing physical speed reducing measures including

roundabouts, road narrowings, chicanes and road humps (Webster, 1993). Such

measures are often backed up by speed limits of 30 km/h, but they can be designed to

achieve various levels of appropriate speed (DRD, 1989, 1991, 1993).

Traffic calming reduces the speed of motor vehicles by various physical modifications:

vertical and horizontal deflections, changes in surface colour and texture, a reduction in

overall carriageway area, and signs and other symbols to convey to drivers that they need

to have greater awareness of vulnerable road users. Gateways may indicate entries into

traffic-calmed areas. Traffic calming measures, based upon various national guidelines,

are now common throughout the EU and are often introduced as part of area-wide urban

safety management Recent experience in the Netherlands has shown positive effects of

traffic calming measures not only implemented in traffic calming areas, but also on

surrounding traffic arteries. Speeds have also been reduced on 'distributor roads' by

constructing traffic calming facilities which include use of roundabouts.

Experience in several EU Member States over the last twenty years has shown that

accident reductions of between 15 and 80 per cent can be achieved by comprehensive

area-wide treatments (Brilon and Blanke, 1993; Herrstedt et al, 1993; IHT, 1990a; CERTU,

1994). The results indicate that application of such speed management measures in urban

areas throughout the EU might reduce the total number of injury accidents by 5%.

hierarchy. Speed limit zones of 30 km/h are most appropriate where an urban safety

management strategy has been adopted. Self-enforcing measures in the zones are usually

necessary to reduce speeds.

Speed limit Risk

km/h

Woonerf and residential roads 30 0.20

Residential roads 50 0.75

Urban arteries 50/70 1.33

SWOV, 1997

As the Table above illustrates, the more urban roads which can more appropriately be

given a ‘residential’ function with the maximum speed limit of 30 km/h the better.

In the Netherlands, re-classification of the road network is well underway to distinguish

between areas where priority can be given to residential, recreational and agricultural

functions, which comprise 65-90 per cent of total road length in the network, and traffic

arteries which give priority to traffic flow. It has been established that two-thirds of the

Dutch road network within built-up areas can be converted into 30km/h zones. Central and

local government have signed an agreement to convert 50% of these streets into 30km/h

zones by the year 2002. To date, the number of kilometres of 30km/h-streets has been

increased to 19,000 km with 9 people killed a year on these streets.

Monitoring of this rapid expansion In the Netherlands over the last few years has shown

very large reductions in casualties, particularly in fatalities - perhaps 17% of fatalities in

urban areas had been saved through this policy by 2000. With more direct conversion

11

from 50km/h to 30km/h having taken place over the last 2-3 years, the impact on fatalities

is expected to increase significantly. There have also been substantial improvements to

the 50/70km/h roads in the Netherlands. These include new roads built to a high standard

incorporating roundabout junctions, improvement of 50km/h roads surrounding 30km/h

zones, and also improvement of some 15% of existing 50km/h roads. The Dutch Transport

Ministry has recently put forward the case to Government for around 2.5 billion Euro billion

up until 2010 for the reshaping of the road network within the sustainable safety

programme (Sunflower Congress, Amsterdam 2002).

In the town of Baden in Austria, about 75% of the road network is today part of a 30kmh

zone or woonerf. Since the introduction of its integrated transport and safety plan in 1988

which introduced a range of measures, the town has seen a 60% reduction in road

casualties (Lines and Machata, 2000).

In terms of cost benefit of area-wide speed and traffic management, research and

experience in the British five towns study (1980s) has shown that the additional travel time,

vehicle wear and tear and fuel costs are small, and reduce the overall benefits by no more

than around 15%.

reducing crashes including collisions with pedestrians through speed limiters to enforce the

posted speed limit. Intelligent Speed Adaptation (ISA) is the global name for systems that

“know” the permitted maximum speed and use that knowledge to inform the driver and/or

intervene in the vehicle’s control to prevent it from being driven faster than the permitted

limit. Intervention control can be by:

this can be overriden by the driver with a “kickdown,”

exceeding the limit, perhaps supplemented by

There are three types of ISA in terms of the degree of intervention of the system. The

information on speed limit is linked to the vehicle controls but the driver can choose

limiting is enforced. Knowledge of the speed limit could come from roadside beacons or

from a modified navigation system in the form of an enhanced on-board digital road map

coded with speed limits for each road combined with a GPS-based location system. The

latter is the so-called autonomous version of ISA which does not require extensive

investment in roadside infrastructure. The most recent estimates of the accident savings

from ISA for all road users have been made by a UK national research project and are

shown in Table 5 (Carsten and Tate, 2000).

These estimates are based on a prediction of 40% compliance with an Advisory system

and 50% compliance with a Driver Select system. Full compliance with speed limits would

occur with a Mandatory system. Clearly, the Mandatory systems predict the largest

accident savings, with the Dynamic Mandatory system being the most effective. These

predictions are broadly in line with estimates previously made for Sweden (Várhelyi, 1996).

It has been estimated that around 20% of pedestrian accidents would be reduced on urban

roads from enforcement of urban speed limits by Mandatory ISA (Carsten and Tate,

2000). Varhelyi (1996) estimated that there would be a 78% savings in pedestrian injury

accidents at pedestrian crossings with an ISA that slowed vehicles to 30 km/h.

12

Fixed 2–21% 4–30% 5–37%

Variable Advisory 2–22% 4–31% 5–39%

Dynamic 3–27% 5–38% 6–47%

Fixed 5–21% 8–30% 10–37%

Driver Select Variable 6–22% 9–31% 11–39%

Dynamic 10–27% 14–38% 18–47%

Fixed 11–31% 15–43% 20–53%

Mandatory Variable 12–33% 17–45% 22–55%

Dynamic 19–50% 28–65% 35–75%

However, a number of steps have to be taken before ISA can be implemented:

1. Agreement needs to be reached on standards for such aspects as: road maps, driver

interface, vehicle control and, for Dynamic ISA, communications. This needs to be

harmonised at a European level to enable a pan-European capability.

2. ISA-capable cars need to be put into manufacture.

3. Before mandatory use can be considered, a majority of the vehicle fleet should be

equipped.

4. There has to be public and political acceptance.

ETSC promotes the need for further research and development towards harmonised

standards for Intelligent Speed Adaptation systems towards an eventual requirement for

ISA capability on all new vehicles sold. In the meantime, encouragement needs to be

given to manufacturers providing ISA systems via the European New Car Assessment

Programme to enable the consumer to start benefiting from a voluntary system and speed

limits need to be introduced into digital road maps.

A description of the range of demonstration projects carried out in Europe is given in the

complementary presentation to this Conference on Priorities for EU motor vehicle safety

design – pedestrian safety.

road. However, while crossings give some protection to the young and elderly, many

crashes occur in their vicinity: the 50m either side of a signalised crossing is particularly

dangerous.

Road lighting, refuges, safety fences and raised pedestrian crossings can all improve the

safety of crossing. A package of measures, including, for example a raised pedestrian

crossing and safety fences, is likely to reduce the number of pedestrian accidents at

pedestrian crossings by about 60% and the number of vehicle accidents by about 35%

(PROMISING, 2001). However, bus stops on refuges in the middle of streets can be

particularly hazardous for pedestrians (OECD, 2001). Where roads are wide and vehicle

flows relatively light, narrowing at pedestrian crossings can be effective.

pedestrian crossings can improve safety especially on higher speed roads or those with

13

high traffic levels (Jensen, 1998). School crossing patrols provide a managed means of

safer crossing for children as a particularly vulnerable group.

Increasingly, where signal-controlled crossings are being upgraded, closer attention is

being given in identifying the pedestrian crossing phase to pedestrian ‘value of time’

assessments which tend to be higher than those of motor vehicle users.

conflict points but should be installed only where there are risks of crashes from

pedestrians walking onto the road. Guardrails restrict people’s freedom and are resented

unless there is no practical alternative. Drivers must be able to see pedestrians waiting to

cross at the end of a length of guardrail.

such as Belgium and the Netherlands, small children are allowed to cycle there. However it

is of much concern to many pedestrians, particularly the elderly and people who are

visually impaired. In specific instances where no on-carriageway solution can be found,

and where visibility is good, it may be appropriate to convert the footway to shared use.

Widening of the footway clear signs and markings will help to make shared use more

acceptable. Segregation by white line only may be expedient but segregation by kerb or

level is preferred by the visually handicapped.

degree of reduced mobility and all of us are sometimes ill, impaired or encumbered. The

resulting needs must be understood before facilities, especially pedestrian crossings, are

the facades of buildings, but they can have problems in finding their way in pedestrian

areas (IHT, 1991). Different surface textures or directional guidance paving can help them.

Street furniture can be a hazard and should not be placed on the natural routes taken by

If steps are unavoidable, the top and bottom of flights of steps should have warning

surfaces. Dropped kerbs at pedestrian crossings assist those with physical impairments

while tactile surfaces help those with visual impairments.

Research has shown that prohibiting on-street parking improves safety. The number of

accidents is reduced by about 25% in streets where on-street parking is prohibited.

About 30% of struck pedestrians fail to see the car before the collision. The more

conspicuous motor vehicles are to road users outside them, and the latter are to drivers,

the more opportunity both will have to avoid collisions. Road layout can help in this and so

can the use of daytime running lights by drivers, the use of lights at night by cyclists, and

the wearing of reflective or light-coloured clothing by pedestrians and cyclists (ETSC,

1999a).

Since the majority of severe pedestrian collisions are with cars, major improvements in

crash protection for pedestrians can be achieved in the short term and with great efficiency

by changing car design.

Developing technical tests suitable for use in legislation to require protection for vulnerable

road users in frontal impacts with cars has been the focus of a 22-year EU-funded

14

research and development programme. Funded by the EU and Member States, the

programme involving national transport laboratories, government departments and

industry, was brought together by the European Enhanced Vehicle-safety Committee

(EEVC).

The pedestrian tests, proposed by EEVC originally in 1991 with an updated report to the

Commission in 1994 (EEVC 1994) and in 1998 (EEVC 1998), are an integrated package of

four tests representing impacts to the parts of the body which most frequently sustain

severe injuries in car to pedestrian impacts. Sub-system tests were used because they

have many advantages over pedestrian dummies for tests intended for legislative use.

The state of the art EEVC tests comprise:

1. Legform to bumper test to prevent serious knee joint injuries and leg fractures

2. Upper legform to bonnet leading edge test to prevent femur and hip fractures and

injuries

3. Child headform to bonnet top test to prevent life-threatening head injuries

4. Adult headform to bonnet top test to prevent life-threatening head injuries

On the basis of national and European studies carried out under the EU programme, it has

been estimated that around 2,000 lives and 17,000 serious injuries to pedestrians and

cyclists could be prevented annually if all cars on EU roads today met these tests. An

updated benefit analysis on pedestrian savings is expected to be published shortly by the

UK TRL.

EEVC-based pedestrian tests have been used since 1996 by the European New Car

Assessment Programme (EuroNCAP) which provides information to consumers on the

crash performance of new cars and which receives substantial Commission funding. No

car tested has yet performed well enough to have passed the EEVC tests proposed for

legislation. Results to date indicate that only 3 EuroNCAP tested cars have received 3 out

of a possible 4 star rating, 65 have obtained two stars and 14 have obtained one star

(Official Report, 2001).

However, just as the European Commission was expected to come forward with

legislation, with a Directive promised in the last two road safety action programmes (, with

a draft proposal for regulation translated into all the Community languages (6065/2000)

and following pressure from the European Parliament and Council of Ministers for a

legislative proposal, the European car industry came forward with an alternative proposal

for a voluntary agreement. The European Commission is currently consulting the

European Council of Ministers and the European Parliament as to whether to accept it or

to propose a Directive (CEC,2001c).

This proposal comprised two phases of pedestrian protection tests (the second phase

15

being subject to review in 2004 before being confirmed) and several other measures

assessed by safety experts to be either peripheral to pedestrian safety or needing

separate treatment by Directive (anti-lock braking, daytime running lights). The detail is

presented in the complementary paper being presented at this Conference.

The Phase 1 tests – the only certain pedestrian sub-system tests in the agreement - have

been roundly criticised by experts as non-scientific (Janssen 2001, Hobbs 2001, Lawrence

2001). The Phase 2 tests mention the possibility of adopting EEVC by 2010 but ‘equivalent

measures’ are allowed and, as previously noted, the second Phase would be subject to a

review in 2004.

The safety content of this agreement has received close scrutiny from experts in the

leading research organisations involved in European pedestrian work and have been

rejected by European non-governmental safety and consumer organisations for several

EEVC tests.

offered a 75% lower level of protection against fatal injury according to the UK TRL

(Official Report of the House of Commons, 12.11.2001). Independent experts involved

in pedestrian protection research told the Commission Hearing on Pedestrian

addition to providing substantially lower levels of protection than the EEVC tests, the

Phase 1 tests were not scientific; the tests were not a natural first step towards EEVC,

could drive car design in the wrong way for effective protection as well as producing

serious side-effects (Janssen 2001, Hobbs 2001, Lawrence 2001).

saved from lower leg fractures would instead suffer serious knee joint injuries, which

are more important because these have a greater risk of permanent disability and

consequently are of higher societal cost.

selected by EEVC and did not represent the adult head, thus providing inappropriate

protection for the adult head and leaving one third of the bonnet area unprotected.

test requirement in Phase 1 would be likely to increase femur and pelvic fractures.

head injuries nor femur and pelvic injury.

The Honda Civic offers now 80% of EEVC (without using new technology) at an

16

additional cost, according to the TRL of only £6.50 (10 Euro) – that is 3 times the level

of the Phase 1 protection which the industry offered to implement fully in 11 years time.

outweighed in a very short time by the large safety gains of a Directive implementing

EEVC.

The opinion of the lead Committee in the European Parliament (four Committees have

considered the issue) has indicated support for the take up of EEVC or equivalent test

methods (which do not exist) by the year 2010 in a Framework Directive. A final opinion is

expected in June and the Commission has indicated that that Parliament’s opinion will be

most important in contributing to their final decision.

ETSC continues to campaign for legislation which implements EEVC with certainty and for

car industry focus on meeting the state of the art EEVC pedestrian tests as soon as

possible. ETSC is also urging the European New Car Assessment Programme to combine

the star ratings from car occupant and pedestrian tests to give consumers a quick

reference guide to the overall crash test performance of new cars. EuroNCAP has

recently taken the decision to continue with EEVC rather than include the voluntary

agreement Phase One testing in its programme.

The attitudes and behaviour of motor vehicle users towards pedestrians are very

important. Training provided by driving instructors, the advice and information that drivers

receive from user and safety organisations, and the influence exerted upon them by

enforcement should all be reoriented to promote attitudes and behaviour based on higher

priority for the safety of pedestrians on the roads the drivers use. Emphasis should be

placed both upon greater consideration and upon greater compliance with traffic laws

concerning speed and giving way, whose effect on the safety of pedestrians is strongest.

Achievement of safe routes for walking and cycling which are also attractive to their

intended users will be helped by consultation with pedestrians, cyclists and prospective

cyclists in the catchment areas of the routes, as well as research into the journeys they

wish to make on foot or bicycle.

Even on the best practicable routes, safer walking calls for competence on the part of the

pedestrians. Information, education and training should therefore be provided for

pedestrians of all ages from the nursery and kindergarten through the school years to

young adulthood, and later as parents and as middle-aged and elderly people adjusting to

the changes in capability that come with advancing years.

Action on pedestrian safety can be taken at international, national and local levels.

Improvements need to be considered within the framework of national and local targeted

road safety programmes and as part of a comprehensive pedestrian safety policy.

Effective implementation of measures for safer walking requires dedicated and technically

informed effort by all of the many professionals involved, together with commitment by

policymakers and the support of a convinced public.

17

This requires systematic dissemination of research-based interdisciplinary technical

guidance that synthesises current best practice to town planners, architects, highway and

traffic engineers, road safety professionals, the police and judiciary, driving instructors,

teachers, those who work with parents and elderly people, and designers of vehicles and

protective equipment. It also requires technically supported guidance in policy formulation

to be communicated to policymakers, who in turn should be encouraged to join with road

safety organisations and road user groups in campaigns to inform the public and win their

acceptance of the necessary policies and measures.

The report on pedestrians (PROMISING, 2001a) described an implementation strategy as

consisting of the following steps:

various levels, for example concerning a whole country or a specific part of a town.

back to the overall understanding of pedestrian safety problems

A better balance between the mobility and safety of all road users is necessary to allow

them to participate fully in society. Walking needs to be recognised as a mode of transport

in its own right if people are to be encouraged to travel on foot or by public transport rather

than by car in order to reduce environmental damage, improve public health, and enhance

the quality of life in towns and cities.

Given that the focus of planning and infrastructure provision for at least the last thirty years

has been to consider the mobility of vehicle users as the main priority, this is clearly going

to take some time.

However, this does not mean that very positive results cannot be achieved in the short

term, whether in infrastructure of vehicle engineering as, indeed the examples set out in

this paper demonstrate.

This paper is based principally on the review activity of the following experts from across

the EU who comprise the following ETSC Working Parties:

Dr Rudolf Gunther (Chairman) (D) Mr Dominique CESARI (F)

Professor Richard Allsop (Editor) (UK) Mr Edgar JANSSEN (NL)

Dr Lars Ekman (S) Mr Anders KULLGREN (S)

Mr Dominque Fleury (F) Prof Klaus LANGWIEDER (D)

Dr Lene Herrstedt (DK) Mr Dietmar OTTE (D)

Dr Christa Michalik (A) Prof. Fernando PINA DA SILVA (P)

Ir Edgar Janssen (NL) Mr Pete THOMAS (UK)

Mr Derek Palmer (UK) Mr Thomas TURBELL (S)

Mr Antiono Lemonde de Maecdo (P) Dr Oliver CARSTEN

(ETSC Road User Behaviour and Telematics

Working Parties) (UK)

I am particularly grateful to Professor Richard Allsop, Chairman of the ETSC Road

Infrastructure Working Party for his help with updating the road safety engineering aspects.

18

European Transport Safety Council, Brussels,1999

Transport Safety Council, Brussels, Belgium, 2001

Co-operation and Development, Paris, France.

European Conference of Ministers of Transport, Council of

Ministers, 30-31 May 2000, CEMT/CM (2000) 2/final

Danish Road Directorate, Copenhagen, Denmark.

report, VTT, Finland, 1998

Final report of the European research project PROMISING,

SWOV Institute for Road Safety Research, Leidschendam,

the Netherlands, 2001

Traffic Planning and Engineering, Lund University,

Sweden, and FACTUM Chaloupka, Praschl & Risser OHG,

Vienna, Austria.

Council, Brussels, 1999

Federal Highway Research Institute (Bast), Bergisch

Gladbach

d’Études sur les Crashes de Piétons, Paris, 12 September.

Institute for Road Safety Research, Leidschendam, the

Netherlands.

COM (2001) 370

EVSC (2000) Final Report External vehicle safety control project,

Institute for Transport Studies, University of Leeds, 2000

Levelt, P., Thomsen J., Thorson O., van Severen J.,

Vansevenant p. Nilsson P., Jorgensen E.,Lund B. &

Laursen J. Danish Road Directorate, 1998

Laboratory report 15, Crowthorne, Berkshire 1993

Prevention, 24(1), 57-65.

18. Crowthorne: Transport Research Laboratory (TRL).

rapport 80. Copenhagen, 1989

19

Copenhagen

7. Copenhagen

The Hague.

HERRSTEDT, L. et al (1993) Herrstedt, L. Kjemtrup, K. , Borges, P., Andersen, P.S.

Copenhagen: Danish Road Directorate, 1993

IHT (1990) Institution of Highways and Transportation.

SWOV Institute for Road Safety Research

Conference, Brussels, Sept 2000.

University of Leeds, UK.

Department of Traffic Planning and Engineering, University

of Lund, Sweden.

Conference, Melbourne, Australia.

December 1998

OFFICIAL REPORT, 2001 House of Commons, 12.11.2001, London

CEC (2001c) Commission of the European Communities

Communication from the Commission to the Council and

the European Parliament - Pedestrian protection:

commitment by the European automobile industry on a

draft negotiated agreement on pedestrian protection

COM(2001)389 final

JANSSEN E. (2001) Test methods to evaluate pedestrian protection. Chairman

- EEVC WG 17 Presentation to Commission hearing on

pedestrian protection, February 6 2001

HOBBS C.A. (2001) Safer car fronts for pedestrians and cyclists. Chairman-

ETSC Vehicle Safety Working Party presentation to

Commission hearing on pedestrian protection, February 6

2001

LAWRENCE G Background to pedestrian protection test methods and

current EU/car industry proposals, Transport Research

Laboratory. Presentation to British MPs, House of

Commons, 17.10. 2001