thematic network...1 thematic network road safety and environmental benefit-cost and...

1

TThheemmaattiiccNNeettwwoorrkk

RRooaadd SSaaffeettyy aanndd EEnnvviirroonnmmeennttaall BBeenneeffiitt--CCoosstt aanndd CCoosstt--EEffffeeccttiivveenneessssAAnnaallyyssiiss ffoorr UUssee iinn DDeecciissiioonn--MMaakkiinngg

EExxaammpplleess ooff aasssseesssseedd rrooaadd ssaaffeettyymmeeaassuurreess

-- aa sshhoorrtt hhaannddbbooookk --

July 2006

Funded by the European Commission

3

RROOSSEEBBUUDD -- RRooaadd SSaaffeettyy aanndd EEnnvviirroonnmmeennttaall BBeenneeffiitt--CCoosstt aannddCCoosstt--EEffffeeccttiivveenneessss AAnnaallyyssiiss ffoorr UUssee iinn DDeecciissiioonn--MMaakkiinngg

EExxaammpplleess ooff aasssseesssseedd rrooaadd ssaaffeettyymmeeaassuurreess

-- aa sshhoorrtt hhaannddbbooookk --

FFeeddeerraall HHiigghhwwaayy RReesseeaarrcchh IInnssttiittuuttee BBAASStt,,GGeerrmmaayy

KKaarrll--JJoosseeff HHööhhnnsscchheeiiddRRoosseemmaarriiee SScchhlleehh

MMaarrkkuuss LLeerrnneerrSSuussaannnnee SScchhöönneebbeecckk

IInnssttiittuuttee ooff TTrraannssppoorrtt EEccoonnoommiiccss TTOOII;; NNoorrwwaayy RRuunnee EEllvviikkKKnnuutt VVeeiisstteenn

IInnssttiittuuttee ffoorr RRooaadd SSaaffeettyy RReesseeaarrcchh,, SSWWOOVV;;NNeetthheerrllaannddss

PPaauull WWeesseemmaannnnCChhaarrlloottttee BBaaxx

KKuurraattoorriiuumm ffüürr SScchhuuttzz uunndd SSiicchheerrhheeiitt KKuuSSSS,,AAuussttrriiaa

MMaarrttiinn WWiinnkkeellbbaauueerrSStteeffaann CChhrriissttiiaannKKllaauuss MMaacchhaattaa

UUnniivveerrssiittyy ooff CCoollooggnnee UUooCC,, GGeerrmmaannyy HHeerrbbeerrtt BBaauummJJuuttttaa SScchhnneeiiddeerr

DDiippaarrttiimmeennttoo IIddrraauulliiccaa TTrraassppoorrttii ee SSttrraaddee RRoommaaDDIITTSS,, IIttaallyy

FFrraanncceessccoo FFiilliippppiiLLuuccaa PPeerrssiiaa

PPiieerrlluuiiggii AAllooiiaa

TTrraannssppoorrttaattiioonn RReesseeaarrcchh IInnssttiittuuttee TTRRII,, IIssrraaeell AAllffrreedd--SShhaalloomm HHaakkkkeerrttVViiccttoorriiaa GGiitteellmmaann

TTrraannssppoorrtt RReesseeaarrcchh LLaabboorraattoorryy TTRRLL,, UUnniitteeddKKiinnggddoomm JJeerreemmyy BBrroouugghhttoonn

CCeennttrree dd''EEttuuddeess TTeecchhnniiqquueess ddee ll''EEqquuiippeemmeenntt dduuSSuudd OOuueesstt CCEETTEE SSOO,, FFrraannccee PPhhiilliippppee LLeejjeeuunnee

IInnssttiittuuttee ffoorr TTrraannssppoorrtt SScciieenncceess LLttdd.. KKTTII,,HHuunnggaarryy PPéétteerr HHoollllóó

TTrraannssppoorrtt RReesseeaarrcchh CCeenntteerr CCDDVV,, CCzzeecchhRReeppuubblliicc

JJaarroossllaavv HHeeiinnrriicchhJJaann TTeeccll

TTeecchhnniiccaall RReesseeaarrcchh CCeennttrree ooff FFiinnllaanndd VVTTTT,,FFiinnllaanndd MMaarrkkoo NNookkkkaallaa

NNaattiioonnaall TTeecchhnniiccaall UUnniivveerrssiittyy ooff AAtthheennss NNTTUUAA,,GGrreeeeccee GGeeoorrggee YYaannnniiss

SSwweeddiisshh RRooaadd aanndd TTrraannssppoorrtt RReesseeaarrcchh IInnssttiittuutteeVVTTII,, SSwweeddeenn HHaannss TThhuulliinn

5

Contents

1 INTRODUCTION ................................................................................................... 7

2 EXAMPLES OF ASSESSED ROAD SAFETY MEASURES – A SHORT HANDBOOK - ....... 8

2.1 HOW TO USE THE HANDBOOK .............................................................................. 8

2.2 USER RELATED MEASURES .............................................................................. 132.2.1 Reforming and improving basic driver training, education and licensing...... 132.2.2 Road safety campaigns................................................................................ 142.2.3 Voluntary training for bus and truck drivers .................................................. 152.2.4 Traffic safety education and information for children and adolescents......... 162.2.5 First aid education and training at school, for drivers and with periodical

repetition....................................................................................................... 172.2.6 Alcohol interlocks ......................................................................................... 182.2.7 Physical examination of vehicle drivers........................................................ 192.2.8 Prohibiting the use of external two-way communication devices in cars and

on motorcycles ............................................................................................. 202.2.9 Reduced alcohol limit for selected groups.................................................... 212.2.10 Disco buses.................................................................................................. 222.2.11 Randomly scheduled and enhanced police enforcement............................. 232.2.12 Random breath testing (RBT)....................................................................... 242.2.13 Speed enforcement ...................................................................................... 252.2.14 Seat belt enforcement .................................................................................. 262.2.15 Section Control ............................................................................................. 272.2.16 Red light cameras......................................................................................... 282.2.17 Combined enforcement and publicity campaigns......................................... 292.2.18 Cycle helmet related campaigns and legislation .......................................... 30

2.3 VEHICLE RELATED MEASURES........................................................................... 312.3.1 Safety inspections of heavy vehicles............................................................ 312.3.2 Automatic tracing of emergency calls........................................................... 322.3.3 Daytime running lights (DRL) ....................................................................... 332.3.4 Pedestrian and bicycle visibility enhancement ............................................. 342.3.5 High mounted and multistage stop lamps .................................................... 352.3.6 Truck visibility enhancement ........................................................................ 362.3.7 Seat belt reminder in passenger cars and ignition interlock for seat belts ... 372.3.8 Underrun guard rails on trucks ..................................................................... 382.3.9 Measures to prevent blind spot accidents with trucks .................................. 392.3.10 Improvement of car front protection to increase the safety of pedestrians and

cyclists.......................................................................................................... 402.3.11 Crash data recorder...................................................................................... 412.3.12 Occupant protection measures for buses..................................................... 422.3.13 Intelligent speed adaptation (ISA) ................................................................ 432.3.14 Adaptive Cruise Control (ACC)..................................................................... 442.3.15 Lane Departure Warning and Lane Change Assistance .............................. 452.3.16 Anti lock braking system (ABS) for motorcycles........................................... 46

6

2.4 INFRASTRUCTURE RELATED MEASURES............................................................ 472.4.1 Variable message signs ............................................................................... 472.4.2 Traffic surveillance and control systems and interference management...... 482.4.3 Road Safety Audits (RSA) ............................................................................ 492.4.4 Black spot management ............................................................................... 502.4.5 Tracks for walking and cycling...................................................................... 512.4.6 Traffic calming .............................................................................................. 522.4.7 Bypass roads................................................................................................ 532.4.8 Signal control at rural junctions .................................................................... 542.4.9 Roundabouts ................................................................................................ 552.4.10 Roadside guard rails .................................................................................... 562.4.11 Middle guard rails and other lane barriers .................................................... 572.4.12 Two plus one roads ...................................................................................... 582.4.13 Redevelopment of railway crossings............................................................ 592.4.14 Road lighting ................................................................................................ 602.4.15 Hazard warning ............................................................................................ 612.4.16 Preventing accidents with animals ............................................................... 622.4.17 Measures regarding skid resistance............................................................. 632.4.18 Winter road maintenance ............................................................................. 642.4.19 Increasing construction site security............................................................. 652.4.20 Accident prevention and performance monitoring: Centralised management

of dangerous goods transport ...................................................................... 66

7

1 Introduction

It is a challenge to develop effective measures for improving road safety –especially when resources are scarce and economic means are limited.Nevertheless, a major target for European policy as well for as for national,regional and local decision makers is to improve road safety significantly.

To improve road safety is not an end in itself – it is an urgent social task. In theyear 2002 about 50.0001 persons were killed in accidents on European roads.In general economic terms the economic loss is very great; for example inGermany it amounts to over 30 billion Euro per year.

In order to help meet this challenge, ROSEBUD was funded by the EuropeanCommission as a thematic network to support users at all levels of government(European Union, national, regional, local) with information about road safetyrelated efficiency assessment solutions. To this end, ROSEBUD broughttogether researchers, policy makers, decision makers and other relevantstakeholders into a co-operative network.

The European Union has the objective of halving the number of deaths onEuropean roads by 2010. To achieve this it will be necessary to implement arange of effective road safety measures to the fullest extent. Reliableknowledge about the effectiveness of road safety measures is a prerequisite forthis task. Efficiency assessment is a vital tool which should help policy-makersto set more effective priorities for road safety measures, and hence lead to aconsiderable reduction of accidents.

One objective of ROSEBUD was to gather relevant experience with road safetyefficiency assessment. This short handbook “Examples of assessed road safetymeasures” is the main outcome of this gathering. A first version of this handbookwas compiled in WP1 and is still available on the ROSEBUD-homepage. The following pages will provide you with an overview should provide you withan overview about a broad variety of road safety measures, their descriptions,interesting examples and significant assessment results.

Most of the measures considered in this document are recently introduced,planned to be introduced soon or under discussion in at least some of theEuropean countries.

Transferability of the results depends on the similarity of the conditions. Thishandbook could only give successful examples but no guarantee thatsuccesses could be repeated everywhere. Some measures address the sameproblem and target group, for example. Their efficiencies depend on theexclusiveness of their implementation.

If you are also interested in the question how efficiency assessments arecarried out a brief chapter will provide you with a summary of themethodological background. Please have a look at the correspondingpublication “Framework for road safety related efficiency assessment” for moredetailed information about methodological questions.

1 In the EU25.

8

This handbook is more dedicated to political decision makers and the generalpublic than to scientists. The handbook tries to be easy to understand andshould support all efforts to increase road safety. The assessments consideredin this handbook have been carried out by renown researchers and institutions.They were published as self-standing publications or in scientific journals. Witheach assessment result the source is given for further studying.

2 Examples of assessed road safety measures – a shorthandbook -

One objective of ROSEBUD was to gather relevant experience with road safetyefficiency assessment. The following short handbook “Examples of assessedroad safety measures” is the main outcome of this gathering. A first version ofthis handbook was compiled in WP1 and is still available on the ROSEBUD-homepage. The following pages will give an overview about a broad variety of road safetymeasures, their descriptions, interesting examples and significant assessmentresults.

Most of the measures considered in this document have been introducedrecently, are planned to be introduced soon or are under discussion in at leastsome European countries.

Transferability of results depends on the similarity of conditions. This handbookcan only give successful examples but there is no guarantee that successescould be repeated everywhere.

Some measures address the same problem and target group. Their efficienciesdepend on the details of their implementation.

The handbook is dedicated to political decision makers and the general publicrather than to scientists. The handbook tries to be easy to understand andshould support all efforts to increase road safety. The assessments consideredin this handbook have been carried out by renowned researchers andinstitutions which were published as self-standing publications or in scientificjournals. With each assessment result the source is given for further studying.

2.1 How to use the handbookThe handbook includes information about various assessed road safetymeasures and efficiency assessment results for these measures. The handbookwas designed to be user-friendly and problem-orientated to enable all potentialusers to use it as a reference base. For easy navigation and to provide a quickoverview, a two step systematic was developed. In the course of the screeningprocess the assessments of road safety measures had already beencategorised by the focus of the road safety measure.

9

User relatedUser relatedmeasuresmeasures

Vehicle relatedVehicle relatedmeasuresmeasures

Infrastructure re-Infrastructure re-lated measureslated measures

Training andTraining andeducationeducation Active SafetyActive Safety Road designRoad design

Traffic LawTraffic LawEnforcementEnforcement

PassivePassiveSafetySafety

RoadRoadconstructionconstruction

Rescue serviceRescue service

AlertAlert

RescueRescue

IncentivesIncentives TelematicsTelematicseSafetyeSafety MaintenanceMaintenance

Accident Preven-Accident Preven-tion and Perfor-tion and Perfor-

mance Monitoringmance Monitoring

PlanningPlanning

FinancingFinancingBudgetingBudgeting

ControllingControlling

Road Safety Road Safety Measures by focusMeasures by focus

Figure 1: Categorisation of road safety measures by focus

The second type of categorisation is the indication of the target group. Importanttarget groups presenting specific road safety problems were selected.

PedestriansPedestrians

BicyclesBicycles

Age Age groupgroup

MotorbikesMotorbikes GoodsGoodsvehiclesvehicles ChildrenChildren

Novice DriverNovice Driver

Road Safety Road Safety Measures by target groupMeasures by target group

ElderlyElderly

Traffic Traffic participationparticipation

CoachesCoaches

Figure 2: Categorisation of road safety measures by target group

Explanatory signs or icons symbolise the focus of the measure and – if aspecial target group is addressed - the target group of the measure in a quickand easy way. For a structured navigation the relevant icons are highlighted.

10

Categorisation by focus of measure

User related measures are measures relating to thebehavioural aspects of the road users

Training and Educatione.g. Campaigns, driver training, retraining courses

Traffic Law and Enforcemente.g. speed and alcohol limits, Safety Belt Wearing Obligation, Increasing fines,Automatic traffic controls, section control

Incentivese.g. disco busses

Vehicle related measures are measures aimed at improvingthe safety level of the vehicles

Active Safetye.g. ABS, ESP, Daytime running lights, Priodic motor vehicle inspections

Passive Safetye.g. Safety belts, Child restraint systems, Helmets, Pedestrian-frendly carfront

Telematics and eSafetye.g. Intelligent speed management, Alcolocks, Safety-belt-interlock-systems

Infrastructure related measures are measures to achieve theaccident reduction potential of current knowledge aboutsafe roads and roadsides

Road designe.g. Safety audits, design guidelines, 2+1 carriageway-roads

Road constructione.g. Road marking, construction site safety,

Maintenancee.g. Management of skid resistance, winter maintenance

Rescue Service refers to measures to improve the treatmentof those who have been injured in accidents

Alerte.g. Standardised international emergency number

Rescuee.g. First-aid training in schools, Periodical repetition of first-aid training

Figure 3: Explanatory signs for highlighting the focus and target group of measures(Part I)

11

Accident Prevention and Performance Monitoring refers tomeasures that improve road safety by optimising theadministration of the road network

Planninge.g. Define targets for road safety work, Guidelines

Financing and Budgetinge.g. Road safety Grants

Controllinge.g.

Measures with negative environmental impact

Categorisation by selected target groupsTraffic participation

Pedestrianse.g. pedestrian friendly car front design, security at pedestrian crossings

Bicyclese.g. helmet wearing campaigns, bicycle lanes

Motorbikese.g. motorbike-ABS, motocycle rider training,

Coachese.g. safety belt wearing obligation in coaches

Goods vehiclese.g. under run protection devices for trucks, measures against blind-spotaccidents

Age groups

Childrene.g. Mobility and Safety education in school

Novice Driverse.g. Graduated licensing, alcohol limit of 0 0/00 for novice drivers

Elderlye.g. physical examination of vehicle drivers

Figure 4: Explanatory signs for highlighting the focus and target group of measures(Part II)

To give a quick overview, the assessment method (CBA or CEA) and the result(poor, acceptable and excellent) of the assessment are highlighted. Themeasures were ranked as

12

B/C-ratioPoor < 1

Acceptable 1 - 3

Excellent > 3

The following compilation of assessed road safety measures includes a shortdescription of the assessed measure, the assessment results for the road safetymeasure and their sources.

Figure 6: Structure of the example pages

The icons on the side ofeach page provide aquick overview of themeasure.

The grey box at the topof each page describesthe road safety measure.

Following the descriptionof the road safetymeasure, an example oradditional information ispresented

The table at the bottomof each page showssignificant assessmentresults

13

2.2 User Related Measures2.2.1 Reforming and improving basic driver training, education and

licensing

Although basic driver training and testing has been widely established, thenumber of accidents caused by novice drivers world-wide is still high. Therefore,different approaches are being discussed to reform and improve national drivertraining, education procedures and licensing systems. For example, basic drivertraining can be improved (e.g. requirement to undertake more practical drivinglessons) and new pre or post licensing measures can be introduced (e.g.license for novice drivers on probation, second phase training after licensing).Moreover, the age at which driver training is permitted could be reduced by oneor two years to allow accompanied driver training in earlier years.

(Source: DVR)

Example from Sweden: The age at which trainingis permitted was reduced from 17,5 years to 16years. The licensing age remained 18 years. Theobjective of this reform was to give novice driversmore opportunities for training before becomingfully licensed. It was assumed that a longer periodbefore passing the driving test would mean thatnovice drivers would be more experienced whenthey took the test and hence have a loweraccident rate.

Selected assessments of improvements in basic driver training,education and licensing B/C-ratio

Reforming basic driver training in Sweden 1.43

Method: CBA Result: acceptable

Source: Elvik R. (2001), Elvik R. (2003), Elvik, R.; Amundsen A.H. (2000)

Introduction of a two-phases model of driving education inSwitzerland 3.50

Method: CBA Result: excellent

Source: VESIPO (2002)

Graduated licensing - lowered age limit for driving training inSweden 1.82


Source: Promising (2001)

The evaluation of PROMISING (2001) found a reduction of 35% in the injuryaccident rate of those who started driver training at 16 compared to those whostarted at 17 ½. The savings in accident costs obtained during the first year ofdriving outweigh the additional costs of training. It seems likely, though, that anyeffect during the second and third year of driving will be smaller than during thefirst year of driving.

14

2.2.2 Road safety campaignsRoad safety campaigns complement other activities aimed at improving roadsafety. Important tasks of road safety communication include raising the publicacceptance for road safety measures (e.g. enforcement measures) anddecreasing the public acceptance of risky behaviours are. Media can influenceattitudes e.g. by informing about rules, explaining consequences of riskybehaviour, informing about police enforcement and about possiblepunishments. Since the target of the campaigns are people or groups of people,and as their behaviour may differ from one country to another, the specificmessages addressed to the target group chosen for a campaign may vary fromcountry to country and even within a single country.

In 1998/1999 for example a roadsafety campaign against drinking anddriving was carried out in 17 Germancounties in 12 Bundesländer. Youngcitizens (females aged 16-24)received a letter and an informationbrochure informing about the risks ofdrinking and driving. Also differentmedia (e.g. local radio) in the countiessupported this road safety campaign.In some counties police enforcementwas also intensified during the roadsafety campaign.

Selected assessments of road safety campaigns B/C-ratio

Road safety campaign against drinking and driving inGermany 4.7


Source: Brilon, W. et al. (2002)

Road safety campaign addressing young road users inSweden 20.0


Source: Trivector Information AB (2002):

Extending ‘speak out’ safety campaign in Norway 16.8


Source: Amundsen, A. H., Elvik, R. and Fridstrøm, L. (1999)

In many cases it is difficult to estimate the impact of the campaign itself,especially if the campaign is combined with elements of enforcement or otherroad safety targeted measures whose effects possibly overlap. Furthermore, theeffects of a campaign have to be separated from general developments in roadsafety.

15

2.2.3 Voluntary training for bus and truck drivers

Standardised safety training procedures for new and experienced drivers couldcontribute the number of accidents with trucks and buses that are caused bydriving errors. Voluntary training courses are being designed to increase thecompetence of professional bus and truck drivers.

The training of bus and truck drivers should include information on vehicleinspections, adjusting safety related equipment, seatbelt wearing, proceduresfor loading and unloading, traffic regulations and driving procedures, overtakinganother vehicle, driving at night, adverse weather conditions, skidding andaquaplaning, what to do in case of an accident and first aid. Furthermore,training courses should include controlled braking exercises at different speedsand on roads of different surfaces, manoeuvring exercises, slalom driving anddriving on ascending slopes as well as learning evasive manoeuvres.

(Source: DVR) (Source: DVR)

Selected assessments of voluntary training of bus and truck drivers B/C-ratio

Training of bus and truck drivers (voluntary) in Norway 2.16


Source: Elvik R. (1999), Elvik R. (2001), Elvik R. (2003)

Apart from the more technical part of the training, behavioural elements such asdefensive driving techniques relating specifically to bus and truck drivers shouldbe stressed throughout the courses.

16

2.2.4 Traffic safety education and information for children andadolescents

Traffic safety education and information aims at ensuring that children andadolescents travel safely to and from school. They should be well prepared fortheir active participation in traffic by walking, riding a bicycle, using a bus or acar as passenger. Therefore, comprehensive education programmes arenecessary to build road safety awareness among school children andadolescents. These programmes should be developed by road safetyprofessionals together with teachers. Furthermore, traffic safety educationshould take place on all school levels to reduce the number of accidents withchildren of all ages.

E.g. in Germany the classes 5 to 10 of the secondary school both the safe andresponsible use of bicycle and moped and the rational choice of the means oftransport are part of the timetable. In the classes 11 to 13 of the secondaryschool there are preferred themes supporting a detailed knowledge towardsscientific questions of phenomena being connected with the modern traffic[physical and technical questions or questions concerning automotive medicine(alcohol, drugs)]. A survey in 2004 indicated that safety education in Germanydiffers greatly from region to region and in a lot of cases there is a lack ofinterest amongst teachers at the upper secondary stage in instruction trafficsafety education.


Selected assessments of traffic safety education andinformation for children on different school levels B/C-ratio

Mobility and safety education on all school levels inSwitzerland 4.3



Older children and adolescents participate more actively in road traffic with theirown vehicles, first by bicycle and later by motorised two wheelers. Therefore,safety education for this agegroups is very important.

17

2.2.5 First aid education and training at school, for drivers and withperiodical repetition

Many victims who are killed in a road accident die in the first minutes after thecrash before the arrival of professional emergency services. First aid actionprovided on the spot in the decisive first minutes can save lives, reduce theseriousness of injuries or at least have a good psychological support value forthe victim and other people involved. As vehicle drivers are often first on theplace of accident it is recommended that they be given first aid training.Consequently, the European Union adopted a Council Directive which makes itcompulsory for every person applying for a driving license to know how tobehave in the event of an accident.

First aid training is already acompulsory part of the drivingeducation in most Europeancountries. Drivers should periodicallyrepeat their first aid training.Moreover, compulsory first aideducation in schools for all childrenshould enhance their later ability togive first aid on the spot.

(Source: DVR)

Selected assessments of first aid education and trainingat school, for drivers and with periodical repetition B/C-ratio

Compulsory first-aid education in schools inSwitzerland 90.0



Periodic repetition of the first aid course every 5years for all vehicle drivers in Switzerland 1.2



18

2.2.6 Alcohol interlocks

Alcohol interlocks (also termed alcolocks) are devices to require the driver totake a breath test before starting the car. If the driver fails the test, the devicelocks the ignition of the car. Recent technical innovations have made alcoholinterlocks largely fraud-resistant. Drivers have to hum while blowing into thedevice or briefly inhale at the end, depending on the type of device. Bothrequirements prevent drivers from using a breath sample that was providedearlier. Drivers could also retested regularly in the course of the journey. Thishelps to avoid fraud at the start of the journey because the sober person doingthe test for a drunk driver would have to stay in the car with that driverthroughout the trip. In principle, alcolocks can be installed as aftermarketdevices in any type of car.

In France a pilot project was started in 2004: drink driving offenders caught witha BAC of 0.8 to 1.6 mg/ml may participate in an alcohol interlock programme asan alternative to prosecution. An alcolock stays in the car for 6 months.Offenders have to pay 1,260 €. The trial has been conducted with 40 offenderswith the goal of including 400 offenders eventually. In Sweden theimplementation is already decided.

Selected assessments of alcohol interlocks B/C-ratio

Alcohol interlock in the Netherlands 4.1Method: CBA Result: excellent

IMMORTAL (2005)

Alcohol interlock in the Czech Republic 1.6Method: CBA Result: acceptable

IMMORTAL (2005)

Alcohol interlock in Norway 4.5Method: CBA Result: excellent

IMMORTAL (2005)

Alcohol interlock in Spain 0.7Method: CBA Result: poor

IMMORTAL (2005)

The examples of IMMORTAL (2005) describe the installation of an alcohol lockin cars for every driver that is caught with a BAC level of 1.3 g/l or higher or fordrivers that are caught twice with a BAC level between 0.5 g/l and 1.3 g/l.Furthermore, there is no accompanying or post project rehabilitation programmefor the driver. Any safety effects after two years, (when alcohol lock is removed)are discounted. Only in Spain are the costs higher than the benefits. The reasonfor this is that the benefits for lives saved are comparatively low for Spain andthe number of “caught” drivers is relatively low (the risk of being caught isestimated to be 25%)

19

2.2.7 Physical examination of vehicle drivers

Especially for elderly drivers medical examinations could be required. Thedriving licensing practice for elderly drivers could include various combinationsof age-dependent requirements for re-licensing such as road tests, medicalreports or vision testing. A physical examination of elderly drivers is analternative to rigorous age limits for driving and should allow people to drive aslong as they can do so safely and effectively. Compulsory periodic testing ofdrivers’ eye sight should reduce the number of crashes caused by drivers withpoor eye sight. If the test shows that the eye-sight is not adequate for driving,the driver should be obliged to wear a seeing aid. This duty should be noted inthe driving license. If the bad eyesight is not correctable, the driving licenseshould be withdrawn.

(Source: DVR)

The study IMMORTAL (2005) assessedeyesight testing in various Europeancountries. The measure consists ofmandatory tests after the age of 45 everytime the license has to be renewed. After theage of 65 every five years a “Useful Field OfView” tests (UFOV – criterion: reduction offield of view should not exceed 40%) isincluded in the standard test. A moderatedecline of the relative accidents risk ratio ledto relatively many lives being saved in theCzech Republic. Although UFOV-testing ledto substantial safety gains, the benefits in theNetherlands and Norway were outweighedby manly the mobility costs.

Selected assessments of physical examination of vehicle drivers B/C-ratio

Mandatory eyesight test for car driver (age >30) inSwitzerland) 1.8



Licensing procedures for older drivers in Australia 0.11 - 0.60

Method: CBA Result: poor

Source: Fildes, Langford, Deery, et al. (1998)

Eyesight Testing in the Netherlands and Norway -20.0 – -4.8


Source: IMMORTAL (2005)

Eyesight Testing in Spain and the Czech Republic 1.5 – 4.0

Method: CBA Result: acceptable -excellent

Source: IMMORTAL (2005)

20

2.2.8 Prohibiting the use of external two-way communication devices incars and on motorcycles

External two-way communication devices in vehicles and on motorbikes can beharmful for road safety. They distract the driver from his driving task if usedwhile driving and consequently increase the risk of an accident. A driver is notsufficienty able to control the vehicle while using a mobile phone and driving atthe same time. To reduce accidents caused by the use of two-waycommunication devices (e.g. handheld mobile phones) during driving, their usecould be banned while the vehicle is in motion or the engine is running.

In the United Kingdomfor example it is illegalto use a hand-heldmobile phone whendriving, even when thedriver stopped attraffic lights or in aqueue of traffic. Thisincludes making orreceiving calls,pictures, textmessaging oraccessing theInternet. The drivermust pull over to asafe location.

(Source: DVR)

Selected assessments of prohibition of the use of external two waycommunication devices in cars and on motorbikes B/C-ratio

Prohibition of the use of external two way communicationdevices in cars and on motorbikes 652.0



21

2.2.9 Reduced alcohol limit for selected groups

A lower alcohol limit for selected groups (e.g. truck drivers) or a trial period(e.g. for novice car drivers) could reduce the number of accidents caused byroad users under the influence of alcohol . Violations should be punished withthe suspension of the driving license. The driving license on probation fornovice drivers is sometimes combined with a period of zero alcohol limit. Areduced alcohol limit would leave no tolerance for drinking and would avoidpeer pressure upon young people to have “just one drink”. Drivers couldmake easier decisions about whether or not to drive. As proposed by theBelgium BOB campaign, groups of young novice drivers could select one oftheir group to act as an alcohol-free driver ("BOB").

(Source: DVR)

The Austrian law for example prescribesa probation period of two years fornovice drivers. In addition, the legalBAC-limit (blood alcohol concentration)for novice drivers was lowered from0.08% to 0.01%. During the probationperiod, the following offences lead tocompulsory participation in a driverimprovement programme as well as toextension of the two years probationperiod by an additional year: Offenceagainst the 0.01%-BAC limit, causing aninjury or fatality, committing a dangerousoffence, for example, seriouslyexceeding the speed limits.

Selected assessments of reduced alcohol limit for selected groups B/C-ratioAlcohol limit of 0,2 0/00 for novice drivers, motorbike riders andtruck drivers in Switzerland 71.0



Lower BAC limit for novice drivers and license on probation inAustria 8.91



Lower BAC limit for novice drivers in the USA 11


Source: Elvik, Vaa (2004)

In the Austrian example, the analysis indicates an accident reduction of 18.7%(number of novice drivers involved in accidents with personal injuries andfatalities related to the number of holders of driving licences on probation.) Thebenefits outweigh the costs by a wide margin.

22

2.2.10 Disco buses

Young male car drivers are at high risk of being involved in a fatal or severeaccident. Their risk is especially high at weekend nights, particularly whenalcohol is involved. It is sometimes argued that they had no alternative to thecar if they wanted to participate in their preferred entertainment at weekends.Disco buses can avoid nighttime accidents involving young drivers,particularly under the influence of alcohol, by offering a safe alternative toyoung people unfit to drive and to young people dependent on taking liftsfrom unfit drivers. The disco bus is an alternative to car use designed to meetthe specific mobility needs of young people.

(Source: DVR)

In order to reduce the highnumber of late-night roadaccidents involving 18 to24-year-olds at weekends,numerous public transportservices, in particular disco-buses, have been set up inGermany since the early1990s.

When introducing discobusses, special care mustbe taken to identify theneeds of young drivers andto achieve a high level ofacceptance in order to runthe system effectively. Asshown below the receivedresults are promising so far.

Selected assessments of disco busses B/C-ratio

Disco buses in Germany 4.06



23

2.2.11 Randomly scheduled and enhanced police enforcement

Traffic law enforcement is a factor that contributes significantly to normativeroad user behaviour and road safety. Randomly selected sites and times of dayfor enforcement achieve good results. Traffic rules are usually enforced bytraffic police forces whose activity and success are generally limited by theresources than can be applied and by established priorities. Nevertheless,many studies have shown that increased police enforcement effort would havea major impact on road safety.

Random Road Watch - a program of randomly scheduled low level policeenforcement has been applied in Queensland since 1991. With 279 policestations involved and 40 road segments per station, the project covered some11,000 RRW sites. The total deployment was some 40,000 hours per year. Thestate-wide effect of the program is equal to a 15% reduction in fatal crashes and8.2% reduction in total reported crashes.

Selected assessments of randomly scheduled speed enforcement B/C-ratio Cost per lifeyear saved

Randomly scheduled low level police enforcement in Australia 55.4


Source: Newstead, Cameron, Leggett (2001)

Selective traffic enforcement programs at high-risk times andlocations in the USA2 $5,200

Method: CEA Result: excellent

Source: Tengs et al. (1995)

Intensified police enforcement in Greece 6.6 to 9.7


Source: ROSEBUD WP4, 2005

Concentrated general enforcement in Israel 3.5 to 5.0



Concentrated general enforcement in Israel 3.5 to 5.0



Tripling stationary speed enforcement in Norway 6.5


Source: Elvik, Vaa, 2004

Tripling drink-driving and seat belt enforcement in Norway 1.2 to 3.6



2 Cost-effectiveness is defined as the net resource costs of an intervention per year of live saved(in 1993 dollars).

24

2.2.12 Random breath testing (RBT)

All EU members have legal BAC limits, mostly of 0.5 mg/ml. Random breathtesting can be conducted by traffic police officers in normal police vehicles andperformed in combination with their other duties or, alternatively, by specialisedpolice groups with special-purpose vehicles. Random means in this context thatthere is no requirement to suspect drunken driving in advance for stopping adriver and carrying out a breath test. Such strategy should aim at substantiallyand permanently increasing the current level of enforcement in the area ofdrink-driving to reach a high probability of being stopped and tested by exposinga large number of road users to unpredictable, well publicised, and highly visibleroadside checks. In most EU countries the police are entitled to use theinstrument of RBT.

(Source: DVR)

Selected assessments of random breath testing B/C-ratio

Increased random breath testing in Norway 1.02 – 1.65


Source: Escape (2001)

Random breath testing in Sweden and Norway 1.50


Source: Elvik R. (1999), Elvik R. (2001), Elvik R. (2003), Elvik, R.;Amundsen A.H. (2000)

Random Breath Testing 36 to 55


Source: ETSC 2003

In the ETSC analysis, two sets of assumptions were made: 3% of all journeys inthe EU are made by drivers exceeding the legal BAC limit and 30% of driverfatalities are over the legal limit; alternatively that 2% of drink-drivers areassociated with 40% of driver fatalities.

25

2.2.13 Speed enforcement

Speed enforcement techniques include mobile and stationary radar, laser speedmeasurement devices, aerial enforcement and mobile patrols. In general,common manual and stationary speed enforcement involves a configurationthat includes an observation unit, typically an unmarked police car more or lesshidden at the roadside, and an apprehension unit comprising of one or moreclearly visible marked police cars. The observation unit is equipped with ameasurement device such as a radar or a laser device and possibly adocumentation device such as a still or video camera. Radar units use highfrequencies to measure the speed of target vehicles approaching or recedingfrom a stationary or moving patrol vehicle while laser instruments use pulses ofinfrared light to measure the speed. Speeding vehicles are detected at the firststation, their description is relayed to the apprehension unit downstream, whichflags them to stop and issues citations to drivers.

(Source: ADAC) Source: BASt)

Selected assessments of speed enforcement B/C-ratio

Speed enforcement in Norway and Sweden 2.89 – 3.62


Source: Elvik R. (1999), Elvik R. (2001), Elvik R. (2003), Elvik, R.; Amundsen A.H. (2000)

Tripling stationary speed enforcement in Norway 6.5


Source: Elvik, Vaa, 2004

Automatic speed control in Sweden 2.98


Source: Andersson G. (2003b)

Current use of automatic speed enforcement (Escape) 2.03 – 8.88

Method: CBA Result: acceptable - excellent


Increasing stationary speed enforcement by a factor of up to six times thecurrent level results in benefits that exceed the costs. The additional benefits ofincreasing speed enforcement from six times to ten times its current level aresmaller than the additional costs. The benefit-cost ratio is calculated with 0.87(welfare economics), 2.13 (value of deterrence) and 7.06 (normative ethics).

26

2.2.14 Seat belt enforcement

Seat belts are intended to reduce the incidence and severity of personal injurieswhen an accident occurs. According to most national traffic laws, car occupantsand all children above a specific age anywhere in the vehicle must be securedby a seat belt (and younger children by special child restraint systems). Lowlevels or decline of safety belt use in some regions are worrying. Enforcementof seat belt laws should raise seat belt wearing rates. Intensive policeenforcement efforts are a major component of seat belt enforcement andeducation programmes designed to reduce avoidable traffic fatalities andinjuries. Visual surveys have often shown that seat belt use has risen followingpolice enforcement campaigns.

(Source: DVR)

According to ETSCestimates, seat belt wearingrates in the in EU varybetween 45% and 95% forfront seat occupants, andbetween 9% and 75% forrear seat passengers(ETSC, 2003). Manycasualties would beavoided if these rates roseas a result of this measure. (Source: DVR)

Selected assessments of mandatory seat belt wearing and seatbelt enforcement B/C-ratio

Increased seat belt enforcement 3.23-3.65



Seat belt enforcement in Norway and Sweden 2.67-3.85

Method: CBA Result: acceptable –excellent


Seat belt enforcement in Norway 3.6


Source: Elvik R. Vaa T. (2004)

Studies have shown that so-called 'blitz' actions, lasting only some weeks, canbe very effective in producing sharp increases in seat belt wearing. To achievelong-term effects, they need to be repeated several times a year.

27

2.2.15 Section Control

Section control systems do not measure speed at a certain point in space andtime, but calculates the average speed by means of passage time in a definedarea. The aim is to force drivers not only to slow down at certain points ofstationary speed control (e.g. automatic speed cameras), but adhere to thespeed limit over the whole distance. It provides live monitoring of traffic flowbehaviour and contributes to harmonising traffic flow performance.

The Kaisermühlen Tunnel is an urban tunnel with separate tubes for eachdirection of traffic. The tunnel provides 3-4 lanes per direction with entrance andexit ramps within the tunnel. More than 90,000 vehicles use this part of themotorway every day (about 10% are Heavy Goods Vehicles.) The aim of thismeasure is to reduce accidents to harmonise of traffic flow (reduction of “Stop-and-Go” traffic or congestion during peak hours). In its first year of operation, areduction in average speed by more than 10 km/h was recorded.

Section Control – Automatic Speed Enforcement in the Kaisermühlen Tunnel (Vienna,A22, motorway)

Selected assessment of automatic speed control B/C-ratio

Section control in the Kaisermühlen Tunnel 5.5



In the assessment of section control in Austria, benefits are reductions in accidents andsavings in road traffic emissions. Running costs of the system are covered by finesfrom speed violators. Capital costs are divided into costs for construction andmaintenance.

28

2.2.16 Red light cameras

Stationary red light cameras collect all of the evidence authorities need toprosecute light-runners at intersections. Usually, a picture of the number plateof the offending vehicle and a picture of the driver’s face are taken. In somecountries, the picture of the number plate is sufficient evidence whereas in othercountries it is not. As the system is usually immobile, a second camera isneeded for the picture of the driver’s face. In a typical red-light system, camerasare positioned on poles at the corners of an intersection, pointing inward tophotograph cars driving through the intersection. Generally, a red-light systemhas cameras at all four corners of an intersection, to photograph cars going indifferent directions and to get pictures from different angles. Red-light systemsuse film or digital cameras. The traffic signals and the triggers of the system areconstantly monitored. If a car sets off a trigger when the light is red, two picturesat the edge and in the middle of the intersection are taken to record theviolation.

(Source: DVR)

A violation occurs when a motorist entersan intersection, often deliberately, sometime after the signal light has turned red. Ared light camera is connected to the trafficsignal and to sensors that monitor trafficflow at the crosswalk or stop line. Thesystem continuously monitors the trafficsignal, and the camera is triggered by anyvehicle entering the intersection above apre-set minimum speed and a specifiedtime after the signal has turned red.

Selected assessments of red light cameras B/C-ratio

Red light cameras at signal controlled junctions and pelicancrossings in Glasgow 2.20


Source: Halcrow Fox (1996)

Red light cameras in Sweden 1.69



Red light cameras in Norway 0.84


Source: Elvik R. Vaa T (2004)

29

2.2.17 Combined enforcement and publicity campaigns

In order to increase the acceptance of a road safety measure by the public andthereby render it more successful, road safety campaigns should be launchedsimultaneously with enforcement programmes. The public should be extensivelyinformed about the road safety problem that is addressed. Mostly, the costs forthe campaign are paid by local or national authorities whereas the control costshave to be borne by the police authorities.


As an example, campaigns and controls of the suitability to drive can becombined. The driving suitability of vehicle drivers, especially when drivingunder the influence of alcohol, drugs and medication, should be controlled aswell as the effect of fatigue and the supervision of the prescribed resting periodsfor truck drivers. These enforcement measures could be accompanied by aninformation campaign. The campaign should start to make drivers aware of theirown suitability for driving.Selected assessments of combined enforcement and publicitycampaigns B/C-ratio

A package of enforcement and publicity aimed at drink-drivingand speeding in New Zealand 7.0


Source: Bliss, Guria, Vulcan, Cameron, (1998)

Road safety television advertising supporting increasedpolice enforcement in Australia 3.9 – 7.9


Source: Cameron, Haworth et al, (1993)

Campaign and controls against dangerous and risky driving inSwitzerland 20.0



Publicity campaigns should make use of as many different kinds of media aspossible to reach a high proportion of the public. Printed media (e.gnewspapers, magazines, flyers) should be involved as well as television orradio.

30

2.2.18 Cycle helmet related campaigns and legislation

Head injury is known to be a major cause of disability and death resulting frombicycle accidents (e.g. fractures of vault or base of skull or intracranial injuries).Increasing bicycle helmet wearing should help to reduce the number of headinjuries. Bicycle helmets contain a thick layer of polystyrene which absorbs theforce of an impact and could reduce the consequences of an accident. Toreduce head injuries to cyclists in all forms of accidents, including thoseinvolving a motor vehicle, every cyclist could be required to wear a bicyclehelmet, and violations would be punished. Additionally, such an obligation couldbe accompanied by an information campaign. Moreover, it is sometimesassumed that campaigns are a potential alternative and could replace anobligation.

(Source: DVR)

Example from Germany: 62% of the Germanpopulation uses a bicycle at least occasionally.Annually, about 600 Germans are killed asbicyclists in road traffic. A little less than 50% of thebicyclists injured in road traffic suffer head injuries.65% of the head injuries occur in regions of thehead that are covered by a helmet. In total, about20% of the fatal and severe injuries may be avoidedby helmet wearing, if all bicyclists wore helmets.The current average helmet wearing rate is 6%(2004) almost constant over the recent years. Thecalculation in ROSEBUD WP4 has shown, that atan increase of the helmet wearing rate up to 26.6%the measure would start to be cost-effective (breakeven helmet wearing rate).

Selected assessments of cycle helmets related campaigns andobligations B/C-ratio

Mandatory wearing of bicycle helmets in Norway 2.7 – 6.2


Source: Promising (2001); Elvik, R., Borger-Mysen, A. and Vaa, T. (1997)

Bicycle helmets, campaign and law in Norway and Sweden 1.30 – 3.09



Compulsory bicycle helmet wearing in Germany andAustria 1.14 - 4.45

Method: CBA Result: acceptable -exellent


31

2.3 Vehicle related measures2.3.1 Safety inspections of heavy vehicles

To reduce the number of crashes attributable to fatigued drivers and thenumber attributable to mechanical defects or unsafe equipment in commercialvehicles, vehicle safety inspections could be undertaken to ensure that vehiclesare well maintained for safe operation. Technical inspections of vehicles shouldbe carried out several years after the first registration, then repeatedperiodically. Licensed inspection technicians should perform these inspections,familiar with all the regulations required for the technical inspection of motorvehicles. Vehicle equipment should be inspected to ensure that all vehiclerelated safety standards and regulations are met. Roadside inspections oftrucks should include checks of the driver’s requirements, the presence ofhazardous materials, the sides and the front of the tractor, the steering axle, allsides of the trailer, brake adjustment, wheels etc. Violations which could bedetected by roadside inspections of trucks comprise for example driver'srecords of duty status violations (e.g. regarding driving hours and rest periods)or technical problems like brakes out of adjustment or inoperable lamps. Theinspections could be carried out by the police or authorised staff.

(Source DVR) (Source DVR)

Selected assessments of motor vehicle inspections B/C-ratio

Roadside inspections for trucks in Sweden andNorway 1.24 - 10.13



Periodic inspection of heavy vehicles in Norway 2.6

Method: CEA Result: acceptable

Source: Elvik R., Vaa T. (2004)

First of all, the commercial advantage gained by road transport companies thatdisobey regulations, especially by having drivers work excessively long workhours, could be reduced by carrying out roadside inspections of trucks.

32

2.3.2 Automatic tracing of emergency calls

One weakness of many national emergency response systems is their inabilityto pinpoint the location of a call, in particular from a cell phone, following anaccident. If the accident victim cannot tell the alert system operator which roadhe had been driving on, an important piece of emergency response informationis missing - the accident locations. A combination of technical solutions could beused to reduce the time between the emergency call and the rescue activity,e.g. technical devices to trace the location of mobile phones sendingemergency calls and automatic emergency calls sent from the vehicle with thehelp of airbag sensors, crash data storage devices and GPS-devices. The in-vehicle eCall is an emergency call generated either manually by vehicleoccupants or automatically via activation of in-vehicle sensors. When activated,this in-vehicle eCall system will establish a voice connection directly with therelevant PSAP (Public Safety Answering Point). At the same time, a minimumset of incident data (MDS) will be sent to the eCall operator receiving the voicecall. The aim is to reduce the rescue time. When medical care for critically andseverely injured people is available at an earlier time after the accident, thedeath rate can be lowered.

(Source: DVR)

Due to a shorter rescue time, eCall will reduce the congestion time, because thefaster arrival of rescue teams, police and towing firms enables the accidentscene to be cleared more quickly. To make the system work efficiently it isessential that a network of PSAP with appropriate technical facilities isestablished.

Selected assessments of automatic trace of emergency calls B/C-ratio

eCall 1.3 to 8.5


Source: Abele, Baum et al 2005

The CBA of Abele and Baum calculated two scenarios (basis year 2002):“pessimistic view” (low success rate and high cost figures) and “optimistic view”(best case)

33

2.3.3 Daytime running lights (DRL)

To reduce the number of daytime multiparty accidents that occur when a driveror rider fails to see another vehicle in time to avoid a collision, some countriesalready require vehicle running lights to be illuminated in daytime. Ancompulsory rule for lights to be switched on when the vehicle is in motion couldbe considered, also a campaign to inform the public. As an alternative to ancompulsory rule for the driver one might consider the equipping all new carswith a device which automatically switches the lights on when the vehicle’sengine is started. Another option is the mandatory installation of dedicateddaytime running lamps which turn on automatically when the ignition is startedand are overridden when regular headlights are activated. Because of theircharacteristics, dedicated daytime running lamps have the advantage ofconsuming less fuel than conventional low beam headlights and therefore leadto lower levels of air pollution.

Example from Austria: DRL is not compulsory in Austria at present, except formopeds and motorcycles. The calculations in ROSEBUD WP4 assume thatDRL would reduce the number of fatalities in daytime multi-party accidents by20%.Selected assessments of daytime running lights on cars B/C-ratio

Mandatory use of Daytime Running Lights (DRL) in Europe 1.24 – 1.80


Source: Koornstra, M., Bijleveld, F. and Hagenzieker, M. (1997), Promising (2001)

DRL Campaign and compulsory rule 7.70



DRL 4.4 – 6.4


Source: ETSC 2003

DRL in Austria 3.6


Source: ROSEBUD WP4

DRL in the Czech Republic 4.3


Source: ROSEBUD WP4

Daytime Running Lights 1.42 – 1.96


Source: Elvik, R.; Christensen, P. and Olsen, S.F. (2003)

If dedicated running lamps are combined with a light sensitive switch theproblem of drivers forgetting to turn on their standard low beam headlightsduring dusk and dawn periods or during periods of poor visibility can beeffectively avoided. ETSC calculated that the additional contribution due toDRL-use for all vehicles to the total costs of pollution (as a result of fuelemissions in road transport) would be 1,0%.

34

2.3.4 Pedestrian and bicycle visibility enhancement

Pedestrians and bicycle riders are sometimes difficult to identify for motorised roadusers. The accident rates of bicyclists and pedestrians could be reduced if morewould be done to make these vulnerable road users as visible as possible. Lightsand reflectors are essential for their visibility. With reflective devices cyclists andpedestrians could improve their own conspicuity and thereby make themselveseasier to be detected and identified by other road users. Every bicycle should beequipped with reflective material of sufficient size and reflectivity to be visible fromboth sides when directly in front of a motor vehicle’s head lamps. Reflectivematerials on clothes could also make pedestrians more visible to motorised roadusers.


When driving in the dark on roads without street lighting, drivers can only see thepart of the road which is lit by the headlights. At the same time, the eye´s ability todiscern contrasts is poorer than in daylight. It is particularly difficult to seepedestrians and cyclists in the dark. Pedestrians, cyclists and people ridingmopeds or motorcycles are much more vulnerable to injury in the event of accidentthan people in cars. The objective of equipping road users and vehicles withreflective materials is to reduce the probability of accidents by increasing visibilityand detection distances.

Selected assessments of pedestrian and bicyclevisibility enhancement B/C-ratio

Improving bicycle conspicuity in Norway > 1



Use of reflective devices by pedestrians inNorway and Sweden 5.09 - 7.58


Source: Elvik R. (1999), Elvik R. (2001), Elvik R. (2003), Elvik, R.; AmundsenA.H. (2000)

35

2.3.5 High mounted and multistage stop lamps

Rear impact crashes often involve pre-impact braking by the leading vehicle.The purpose of high mounted stop lamps is to safeguard vehicles from beingstruck in the rear by another vehicle. When brakes are applied, the highmounted stop lamp warns drivers of following vehicles that they have to slowdown. Today, high mounted stop lamps are required in a growing number ofcountries world-wide. These stop lamps, which are installed higher and midwaybetween the rear brake lights to form a triangular pattern, have become astandard equipment on new generation vehicles. Multistage stop lamps have adifferentiated light intensity depending on the brake pedal power. The higher thebrake power on the pedal, the brighter the stop lamps will shine. Because of thedifferentiated light signal, the following drivers can react faster and can assessthe brake power more accurately.

There are different varieties of multistagestop lamps. With some of them the shiningarea is enlarged, with others the stop lampsstart flickering.(Source: DVR)

Selected assessments of high mounted and multistage stoplamps B/C-ratio

Centre High Mounted Stop Lamps (CHMSL) in passengercars and light trucks in the USA 3.18


Source: Kahane, C.J. (1998).

Installation of high mounted stop lamps in Norway andSweden 3.89 – 9.07



Installation of multistage stop lamps in vehicles inSwitzerland 3.40



A normal brake lamp is usually only visible to the car immediately behind thatcar in a queue. When driving in queues, this means that the driver’s delayedreaction spreads backwards through the queue, and the available reaction timebecomes shorter and shorter for each car, the further back it is in the queue.The use of daytime running lights may also make stop lamps more difficult tonotice (on some cars taillights and stop lamps are located in the same place butshine with different intensities). Installing extra high-mounted stop lamps cancounteract these problems.

36

2.3.6 Truck visibility enhancement

Retro-reflective material could increase the conspicuity of trucks at nights and inbad weather. The sides and rear of the trailers can be equipped with retro-reflective tape or reflex reflectors to reduce side and rear impacts into heavytrucks,. the retro-reflective material brightly reflects other motorists` headlights,especially in the dark, and warns them that they are closing on a heavy trailer.Additionally, it is easier for the other road users to assess the distance to andthe speed of the truck. Because of the enhanced visibility of heavy trucks andtrailers, a large-scale introduction of retro-reflecting contour marking couldreduce the probability of side and rear impacts by other vehicles.

(Source: DVR)

SWOV studied the road safety effects,economic efficiency and possibilities of large-scale introduction (voluntary or compulsory) ofretro-reflecting contour marking on lorries. Eachyear there are about 9 deaths and 83 inpatients resulting from collisions with the flankor rear of lorries during twilight and night-timehours. It is estimated that the completeintroduction of retro-reflecting contour markingwould reduce this by 2-3 deaths and 20-30 inpatients per year.

Selected assessments of truck visibility enhancement B/C-ratio

Retro-reflecting contour marking on lorries in the Netherlands > 1


Source: SWOV (2002)

Enhancing the visibility of heavy trucks in Switzerland 1.7



37

2.3.7 Seat belt reminder in passenger cars and ignition interlock for seatbelts

New cars could be equipped with a seat belt-ignition-interlock system to reduceinjuries of car occupants involved in a crash by increasing the seat belt wearingrate. In cars with seat belt-ignition-interlocks the ignition of the engine is onlypossible when all occupants have fastened their seat belts. Alternatively,several devices have been developed to remind vehicle occupants to buckle up.The information could be transmitted by symbol-only reminder systems, flashinglights on the dashboard or a warning tone of reasonable intensity. An audibleseat belt reminder gives a warning tone whenever a seat is occupied but theseat belt is not fastened. A simple continuous reminder is a device that gives awarning as long as the seat belt is not worn, but it is not designed with anignition interlock function.

It is estimated that an audible seat beltreminder for the front seats can raise seatbelt wearing among front seat occupants upto 97%. Based on a review of evaluationstudies it could be assumed that wearing aseatbelt reduces the fatality rate by 50% forfront seat occupants.

Selected assessments of seat belt reminder in passenger cars andignition interlock for seat belts B/C-ratio

Seat belt reminder in passenger cars in Sweden 11.34 - 11.85



Ignition interlock for seat belts in Norway and Sweden 4.40 – 28.36


Source: Elvik R. (1999), Elvik R. (2001), Elvik R. (2003), Elvik, R.; Amundsen A.H. (2000)

Compulsory seat belt-ignition-interlock-systems inSwitzerland 1.10



Audible seat belt reminder 6


Source: ETSC 2003

The European New Car Assessment Program (Euro-NCAP) will provide addedpoint bonuses for cars they assess for crashworthiness if vehicles are fitted withseat belt reminders. In the ETSC study each country of the EU15 has beengiven a weight (proportional to the number of passenger cars registered in thatcountry), used to compute the average wearing rate for seat belts (meanwearing rate in 2000: 76%).

38

2.3.8 Underrun guard rails on trucks

Many motorised road users are killed due to collisions of their vehicles with therear end or sides of trucks that are not equipped with adequate underrunguards. When a car collides with the rear end or side of a truck it could continueto drive under the chassis of the truck. The invasion of the passengercompartment that follows the underriding could have horrible consequences forthe car occupants. Car occupants could be decapitated or cause severe headand upper body injuries. An injury reduction for road users involved in a crashwith a truck could be reached by providing guard rails around all sides of trucksto prevent underriding and to make the crash less severe. Therefore, thecompulsory equipment of trucks with comprehensive side, front and rearunderrun protection devices can be recommended.

Source: DVR

Swedish laboratory tests have shownthat underrun guard rails significantlyreduce the risk of injury, especiallycombined with the use of seat belts bycar occupants.

The objective of side under-runprotection is mainly to preventpedestrians and riders of two-wheeledvehicles from being run over, by gettingcaught in the open space between thewheel axles on large vehicles. Sideunder-run protection can also preventsmaller cars from driving under orbetween pairs of wheels on largervehicles.

Selected assessments of underrun guard rails on trucks B/C-ratio

Front, side and rear underrun guard rails on trucks inNorway > 1



Improving underrun guard rails on trucks in Norway 1.18



Comprehensive underrun protection devices fortrucks in Switzerland 4.10



Under-run guard rails on trucks in Norway 3.9



39

2.3.9 Measures to prevent blind spot accidents with trucks

A truck driver must use side mirrors to see what is happening behind him, but atruck's rear-view and side mirrors are not always sufficient and "blind spots" arecreated. Large trucks have those blind spots located around the front, back, andsides. Common blind spots for a truck driver exist near the right front wheel ofthe truck and at the rear of the trailer. When other road users are in a blind spot,the truck driver is unable to see them. For example, in passing close behind atruck which is preparing to back up, a car would enter the truck's blind spot andan accident may occur. Several vehicle-related, infrastructure-related,organizational and educational measures could help to reduce the number ofblind-spot accidents with trucks, particularly involving cyclists. These includeimprovements at traffic intersections and the installation of special mirrors intrucks; supporting thematic campaigns could be considered.

Efforts in Switzerland to preventblind-spot accidents between cyclistsand trucks at intersections include:infrastructural improvements of trafficintersections, installation of TRIXI-mirrors at traffic intersections,installation of special TOWISPICK-mirrors in trucks, informationcampaigns for truck drivers,information and educationcampaigns for children.(Source: DVR)

Selected assessments of measures to prevent blind spotaccidents with trucks B/C-ratio

Measures to prevent blind-spot accidents with trucks inSwitzerland 1.4



In Denmark, an extra close-up mirror and a wide-angle mirror on the right sidehave been compulsory since 1988. The measure was introduced to reduceaccidents where cyclists ride under lorries when these are turning right atintersections. Studies have shown a tendency for the number of injury accidentsto increase and the number of fatal accidents to go down. None of the changeswere statistically significant (Elvik 2004). It was found that more than half of theclose-up and wide-angled mirrors were wrongly installed.

40

2.3.10 Improvement of car front protection to increase the safety ofpedestrians and cyclists

Pedestrians are mainly killed in accidents with cars. An improvement to thedesign of car fronts can reduce the number of casualties and fatalitiesassociated with collisions with car front-ends. Softer and more flexible bumpersand bonnets could save lives. On the contrary, external features such as bullbars pose an obvious danger. Stronger rules for frontal protection devices andfront and back spoilers could be recommended, demanding that these devicesdo not cause any further damage to a crash victim than the standard vehiclemodel.

(Source: EuroNCAP)

Example from the Netherlands: It isestimated that the annual numberof casualties could be reduced by750 (including 11 fatalities and 263hospitalised) as a result ofimproving the car front-end design.A positive benefit-cost ratio may beattained at a cost of up to 50 ECUper new car assuming that eachyear some 500,000 new carsreplace the same number of oldercars. A positive ratio of benefits tocosts of 3:1 is possible.

Selected assessments of improvement of car front protection B/C-ratio

Improvement of car front design to increase crash-safety ofpedestrians and cyclists in the Netherlands 3.0


Source: van Kampen, Ir L.T.B. (1994)

Tightening the law for front protection devices in Switzerland 150.0



New safety standards for front and bumper in Sweden andNorway 4.66-6.80



The Swiss analysis consists of costs for public relations and for ex-postevaluation of the measures impact on road safety. Furthermore, it is estimatedthat savings from the removal of frontal protection devices (bull bars andspoilers) are nearly equal to the retrofitting costs. The whole average cost isestimated with 120,000 CHF per year (considering a 10 years implementationperiod.) The estimated benefits exceed the costs by a wide margin.

41

2.3.11 Crash data recorder

A crash or event data recorder is an on-board device capable of monitoring,recording, displaying or transmitting pre-crash, crash, and post-crash dataparameters from a vehicle. All vehicles could be equipped with such ancompulsory, crash-resistant and non-manipulable digital device to record thespeed profile and other relevant vehicle operation data. A data recorder caninform traffic accident investigators about, for example, the vehicle’s speed,pressure on the accelerator pedal and application of the brakes. Thisinformation is collected directly before the impact. Crash data recorders areexpected to contribute to road safety in two ways. On the one hand, thesequence of events preceding crashes could be clarifed. Statements on thecause of the crash and conclusions drawn (e.g. if there was a chance to avoidthe crash) can be made quickly and qualified. These are important advantagesfrom a juridical point of view. On the other hand, crash data recorders couldalso have a preventive effect for the drivers and thus improve road safety ingeneral.

1 2 3 4 5 6 7 8 9 1 0 1 1 1 2R 1

0

2

4

6

8

1 0

Crash Data Recorder

Selected assessments of crash data recorder B/C-ratio

Crash data recorder in Sweden and Norway 1.11 – 1.50



The crash data recorder or event data recorder (EDR) is sometimes referred toas the "black box". EDRs can be simple or complex in design, scope, andreach. In some ways similar to data recorders used on airplanes and trains, thecar's recorder springs into action as part of the air bag system. Event datarecorder can tell traffic accident investigators about the car's speed, engineRPMs, how far the accelerator pedal was pressed, if the brakes were applied,whether the driver’s seatbelt was buckled and what warning lights were on - allfrom about five seconds before impact. When an air bag deploys, the data isrecorded onto a computer chip. The data can be retrieved and presented in areport.

42

2.3.12 Occupant protection measures for buses

Several measures are under discussion to prevent injuries to bus occupants,not at least because bus collisions generally receive considerable mediaattention and public focus. These measures include techniques like seatspacing, seat back padding and seat belt systems. Often, primary focus isplaced on school bus safety. In the discussion of passenger seat belts inschool buses it is argued that children could be killed or injured becausethere are no restraint devices and without these devices the safety standardsare not sufficiently effective. Seat belts could offer superior protection in thecase of rollovers, side impact or angle collision of school buses. On the otherhand, in a head-on collision a belted child in a school bus could receiveworse head injuries than an unbelted child. Consequently, seat belts might domore harm than good in a frontal bus collision.

The dimensions of the vehicle itself provideoccupants of heavy vehicles with greater protectionagainst injury in accidents than occupants ofsmaller vehicles.


Selected assessments of occupant protection measuresfor buses

Cost per lifeyear saved B/C-ratio

Seat belts for passengers in school buses in the USA $2,800,000

Method: CEA Result: poor

Source: Tengs et al. (1995)

Seat belts for passengers in buses in Norway 0.02Method: CEA Result: poor


In Norway, to equip one bus with seat belts create costs of NOK 16,500 (ElvikR., Vaa, T. 2004). This is far more than the benefit of injuries prevented. Even ifseat belts could eliminate all injuries to bus passenger the measure would notbe cost-effective.

43

2.3.13 Intelligent speed adaptation (ISA)

Intelligent speed adaptation (ISA) is a system of in-vehicle speed limitation. ISArequires that the location of vehicles be known accurately so that they can beprovided with the local speed limit. This can be achieved through a combinationof a Global Positioning System (GPS) and digital road maps. Furthermore, theelements of the vehicle's power train like throttle, ignition, fuelling system,gearbox and brakes have to be linked. Dynamic intelligent speed adaptationallows temporary speed limits because of specific network or weatherconditions such as slow traffic in fog or near schools at the beginning and endof the day. Mandatory dynamic ISA promises to be the most effective of the ISAsystems in terms of casualty reduction. ISA could lead to enormous reductionsof injury accidents and fatal crashes. The reduction in fatal crashes could beparticularly high because ISA is targeting speeding risks by ensuring completecompliance with the speed limits. ISA is unable to function as well without anaccurate digital road map showing speed limits.

Example from the UK: All the benefitto cost ratios exceeded 5.0 withmandatory ISA showing the clearestbenefits (up to a ratio of 16.7 under ahigh GDP growth scenario).Mandatory ISA is clearly the mosteffective system. If variable anddynamic facilities are incorporatedinto the mandatory ISA, in order tomake the speed limits more up-to-date both spatially and in terms oftime (responding to current weatherconditions for example), thenaccident savings are almost doubled.

Selected assessments of intelligent speed adaptation B/C-ratio

Intelligent speed adaptation in UK >5.00


Source: Carsten, O.; Tate, F. (2001)

Intelligent speed adaptation devices in Sweden 1.37



Intelligent speed adaptation is distinct from the adaptive cruise control (ACC)systems, fitted to many new vehicles.

44

2.3.14 Adaptive Cruise Control (ACC)

Adaptive Cruise Control will enable the vehicle to maintain a driver-defineddistance from the preceding vehicle while driving within a maximum speed limitset by the driver. Since the system only functions at speeds between 30 km/hand 200 km/h, it is designed primarily for use on motorways and rural roads. If,however, there is a rapid reduction in the vehicle’s speed, the system will warnthe driver and switch off to let the driver assume control.

The time gains attainable using ACC can reduce rear-end collisions and caninfluence the severity of those accidents which cannot be avoided (lower vehiclespeed and crash impact, leading to a reduction in accident severity.)

100

30

180

Speed in km/h

Selected assessments of ACC B/C-ratio

Safe Following - ACC 0.9 to 1.2


Source: Aberle, Baum et al 2005

The calculations are based on market analyses: 3% of all cars in the EU25 willbe ACC-equipped by the year 2010 (in 2020: 8%). The good result of 1.2 isreached in 2020.

45

2.3.15 Lane Departure Warning and Lane Change Assistance

Lane Departure Warning (LDW) systems assist drivers in keeping to their lanesby warning drivers when their cars are in danger of leaving their laneunintentionally (mainly due to lack of driver attention). Current systems useeither an audible beep or a “rumble strips” noise, which mimics the sound madewithin a tyre runs over al lane divider. Lane Change Assistants (LCA) assistdrivers intending to change lanes. The LCA monitors the adjacent lanes andwarns the driver if another vehicle is likely to com within colliding distanceduring the lane change. This occurs, for example if the other vehicle is locatedin the LCA-equipped vehicle’s blind spot. LDW can avoid or reduce the severityof accidents in which two vehicles collide frontally (head-on collision) and inaccidents in which a vehicle leaves the road without colliding with anothervehicle.

LDW warning enables a driver to react, on average, 0,5 seconds earlier thanwithout the system. This effects a collision reduction of 25% of all relevantaccidents. The time gain of LCA is 0,7 seconds. This leads to expect a 60%reduction of the number of relevant accidents.

(Source: BASt)

Selected assessments of LDW and LCA B/C-ratio

LDW and LCA 2.0 - 2.1


Source: Aberle, Baum et al. 2005

In the CBA of Abele et al. (2005) both systems have been analysed as a singlecombined system. 0,6% of all cars in the EU25 will be equipped with LDW andLCA by the year 2010 (in 2020: 7%).

46

2.3.16 Anti lock braking system (ABS) for motorcycles

Anti-lock braking systems compensate for rider error in emergency situations.The reflex of emergency braking on a motorcycle usually leads to locking one orboth wheels which immediately creates a high risk of falling off the vehicle; thisis prevented by the use of ABS. Generally it will enable motorcycle riders toimprove their braking performance significantly. On the one hand, ABS is highlybeneficial in reducing motorcycle accident numbers and severity. On the otherhand, ABS is relatively expensive and still not very popular among motorcycleriders, mostly due to the high costs. From the traffic safety point of view,measures must be taken to encourage fitting of ABS to motorcycles, i.e. to raiseconsumers' willingness to invest in ABS.

(Source: DVR)

For the motorcycle, the reflex ofemergency braking usually leads tolocking one or both wheels, whichimmediately creates a high risk offalling off the vehicle. Motorcycleriders are well aware of this dangerand leave a huge "safety gap"between the decelerations theyactually apply and the realdecelerating potential of theirvehicles. Motorcycle drivers use onlyabout 60% of the decelerationpotential of their vehicles.

Selected assessments of ABS for motorcycles B/C-ratio

ABS for motorcycles (with and without taxreduction)

1.11 to 11.73

Method: CBA Result:acceptable to

excellent


The ROSEBUD study considered a crash reduction potential of min 8% andmax 10%. In the analysis of this measure the "Normverbrauchsabgabe"("NoVA"), which can be translated as "fuel consumption tax" is considered. Formotorcycles, this tax equals 0.02% of the net price multiplied by the capacity incubic centimetres, then reduced by 100. It was intended to discount the value ofABS from the NoVA assessment base.

47

2.4 Infrastructure Related Measures2.4.1 Variable message signs

Variable message signs are infrastructure related facilities which could supplytraffic related and safety relevant information to the road user. Variablemessage signs include simple prism displays as well as fully graphic-enableddisplay boards. For example, traffic signs, lane signals or textual displays canbe brought rapidly to the attention of the road user. The importance of variablemessage signs for road safety depends on the messages that they convey.Variable message signs can have at least a short term effect on drivingbehaviour but messages must not be overloaded.

(Source: DVR)

Example Finland: Two variable speed signalzones consist of weather observation points,which are connected by data transfersystems to the computer that sets the speedlimits according to the weather conditions.The projects have been implemented despitetheir economically unconvincing startingpoint. However, there are signs that theseprojects bear additional unforeseen benefits,such as those deriving from increasing theknowledge of the road authority for carryingout similar arrangements in future.

Selected assessments of variable message signs B/C-ratio

Variable speed control sign in Finland <1.00


Source: Nokkala, M; A. Schirokoff (2001); Lähesmaa, J. (1997)

Variable message signs in Sweden and Norway 1.13 - 1.45



Variable message signs in Norway 0,65



The analysis of Elvik and Vaa (2004) consists of a variable speed limit sign at aschool. It is assumed that the road past the school has an annual average dailytraffic of 5,000 vehicles and 0.4 injury accidents per million vehicle kilometres.The saving in accident costs are estimated to be NOK 579,000, increased timecosts to be NOK 598,000 and decreased environmental costs of NOK 0.01kroner per vehicle kilometre. The total benefit of NOK 39,000 are lower than thetotal costs of NOK 60,000.

48

2.4.2 Traffic surveillance and control systems and interferencemanagement

Traffic surveillance and control systems are expected to provide incidentmanagement, speed control, queue detection and warning, lane control and co-ordination with the Urban Traffic Control system and other agencies. The mainreasons for the introduction of traffic surveillance and control systemdevelopment are mostly related to capacity and demand management and notto road safety. Nevertheless, these systems have an effect on road safetythrough preventing or warning of dangerous situations as well as throughoptimisation of emergency services. Traffic interference managementcomprises the localisation and the assessment of traffic interference as well asthe selection of countermeasures which have to be introduced to return to thenormal condition. Modern traffic interference management carried out in controlcentres could comprise automatic devices to detect incidents like fires, trafficaccidents or wrong-way drivers as well as automatic lane and tunnel monitoringsystems.

Example from Israel: The Ayalon freeway, which links the Tel Aviv MetropolitanArea and the Central Business District as well as being an essential North-South route between Jerusalem and Haifa, was extended in 1991. The roadthen began to suffer from regular congestion. New control strategies, includingspeed control and queue detection and warning, were expected to reduce bothprimary and secondary accidents. Also, direct communication with a trafficcontrol centre promised a quicker call for emergency services which couldimprove the survival rate of serious injury. The introduction of the trafficsurveillance and control system for the urban highway proved to be beneficial.Cost savings due to safety effects were estimated at 20% of the total benefits.

Selected assessments of traffic surveillance and control systems B/C-ratio

Traffic surveillance and control system for an urban freewayin Israel 1.7 – 6.3


Source: Kelly, Blum (1992)

The evaluation considered future disbenefits of current operation mode (nocontrol), and savings in these disbenefits which can be attained due to theintroduction of a Control System. The estimates were prepared for a 20 yearperiod, with 8% discount rate. The disbenefits included: accidents, incidents,congestion, maintenance and off-ramp queues' costs.

49

2.4.3 Road Safety Audits (RSA)

A road safety audit is a systematic examination of the safety standard of a road,usually at the design stage, but sometimes also immediately before the road isopened to traffic. The idea of the road safety audit was first developed in GreatBritain and is applied now in many other countries. The two main potentialbenefits from the Road Safety Audit process are to reduce the frequency ofaccidents and casualties and to reduce the need to redesign a scheme after ithas been implemented. Audits are performed by independent auditors and arebased on detailed checklists listing the items to be examined. The independentauditor or an auditor team is commissioned by the owner of the infrastructure(federal, regional and local authority, private owner). The Auditor should haveexperiences with road safety and construction. Road safety audits are oftendescribed as a first step to implement a complete quality management systemfor roads. The aim of the safety audit is to put a value, from the standpoint oftraffic safety, on all new road construction projects and major road maintenanceworks on existing roads, so that any shortcomings in road safety could bedetected in time.

Before RSA After RSA An intersection inGrand Rapids,Michigan: Thetraffic signals arenow hung on abox span of wireand they are nowable to be hungdirectly over thetravel lanes.

Now there are three traffic signal heads, two for the through lane and one forthe left turn lane. Pavement markings now show a sepa-rate left-turn lane atthe intersection.

Selected assessments of road safety audits B/C-ratio

Safety Audit - Denmark 1.46


Source: Herrstedt L. (1999); Herrstedt L. (2000)

Implementation of Road Safety Audits (RSA) in Germany 4-99


Source: BASt (2002)

Road Safety Audits in Norway 1.34



The effects of road safety audits depend on the application of the proposalsmade by the auditor. The effectiveness of road safety auditing is a “derivedeffectiveness” – depending on the effectiveness of the implementation of theproposed measures.

50

2.4.4 Black spot management

“Black spot” is a colloquial term for points, sections or junctions in the roadnetwork that show a regional higher-than-average density of fatalities andsevere injuries. Black spot treatment makes use of the accident record andother information to identify the measures that are likely to be most effective;these are then implemented in those places where they should have thegreatest effect on accidents. Black spot treatment is an iterative procedure witha regional scope based on accident reporting, accident analysis, data storageand presentation. In detail, black spots are identified with the help of informationon the precise accident location and on the accident scenario of fatal andsevere accidents – displayed on a map. Experts on road safety should then visitthose sites, discuss and identify remedial measures. As to the type of measure,low cost measures can be distinguished from major repair works or even areconstruction for the medium to long run. For each measure an accidentreduction potential should be identified and costs estimated. With these data, abenefit cost ratio is calculated. A ranking of the measures for the entire regionaccording to the benefit cost ratio and the time frame, will then allow to define awork programme.

Example from Norway: 62 Sites with most road accidents in Hamar Municipalitywere selected. The number of reported accidents was used as the selectioncriterion. The sites were inspected by representatives of the local office of theNational Road Administration, the municipality, the police, vehicle inspectorateand others. At each inspected site measures were proposed. Two separate costassessments were made for the measures that could be taken immediately andfor some of the "future" measures. The types of measures taken were: erectionof road signs, imposition of Give Way rules, construction of refuges, moving oftraffic signal posts, erection of barriers, marking out of traffic lanes, designationof separate lanes for turning traffic, raising the level of pedestrian crossings,reduction of the sizes of intersections, improvements to visibility and roadmaintenance. Selected assessments of black spot treatment B/C-ratio

Black spot treatment program in Australia 4.10 – 5.10


Source: Newstead, Corben (2001)

Measures at accident black spots in Hamar municipality,Norway 35.00


Source: Statens Vegvesen, Hamar kommune (1993)

Redevelopment of locations with large numbers of trafficaccidents in Switzerland 13.00



The performance of black spot programs relies heavily on systematic andrigorous methods for identifying hazardous locations and on implementingappropriate treatment types that target predominant casualty crash types.

51

2.4.5 Tracks for walking and cycling

Many pedestrians and cyclists do not feel safe in traffic, especially when theyare travelling in mixed traffic on roads with heavy car traffic. Tracks for walkingand cycling, together with footbridges and pedestrian tunnels, are intended toseparate pedestrians and cyclists physically from motorised traffic. Anotherobjective is to give pedestrians and cyclists increased mobility and feeling ofsecurity when travelling in public traffic areas.

Many cyclists are killed as aconsequence of conflicts between theirown road use and the road use of cars.Specific lanes for bicycle riders couldbe useful to reduce the number ofaccidents of these vulnerable roadusers. Bicycle lanes can and should beused to improve riding conditions forbicyclists as well as protecting bicyclistsby separating them from vehicle traffic.On bicycle lanes cyclists can rideundisturbed from motor vehicles orpedestrians. Bicycle lanes help to betterorganise the flow of traffic and reducethe chance that motorists will stray intocyclists’ path of travel. Moreover, lanesand paths for vulnerable road userscould be combined to a whole networkto create higher benefits than isolatedlanes.

(Source: DVR)

Selected assessments of bicycle lanes and pedestrian andbicycle networks B/C-ratio

Bicycle lanes in urban areas in Norway 9.74



Pedestrian tunnel in Norway 4.2


Source: Elvik and Vaa (2004)

The evaluation of the safety of a pedestrian tunnel in Norway considered anannual average daily traffic of 6,000 motor vehicles. It is further assumed thatthere are 0.05 pedestrian accidents per million vehicles passing and 0.05accidents involving vehicles only per million vehicles passing. The costs of themeasure is estimated with NOK 0.6 million. The benefits of NOK 2.5 millionconsists of a reduction of 80% of accidents involving pedestrians and a 10%reduction of accidents with involving vehicles only.

52

2.4.6 Traffic calming

Traffic calming has been widely implemented in Europe. The purpose of trafficcalming is to improve traffic safety for vulnerable road users. Traffic calming isthe combination of mainly physical measures to alter driver behaviour andimprove conditions for vulnerable road users. Driving speed has a majorinfluence on the probability of becoming involved in an accident and on theseverity of injuries. A common problem for example in school zones isexcessive vehicle speed and traffic volume in areas where children must crossstreets and where they are picked up and dropped off. To avoid conflicts, trafficcalming devices should be simple, self-enforcing and easy to modify toaccommodate emergency and other service vehicles. Speed humps arefrequently chosen as a typical solution when there is a need to reduce travelspeeds on a local street and to provide the street with a calmer and safercharacter. The main advantage of speed humps are in their self-enforcingnature and in creating a visual impression that the street is not designated forhigh speeds or for passing traffic. The length is usually larger than the distancebetween the wheels of vehicle (usual length 3.6 m), their height oscillatesbetween 7.5 – 10 cm and the recommended distance between successivehumps varies from 60m to 100m.

Example from Greece: In Athens a limited number oftraffic calming measures has been constructed. TheMunicipality of Neo Psychiko is the only area in theGreater Athens Area which inaugurated anextensive road traffic calming programme at thebeginning of 1990´s. The measures mainly includespeed humps and woonerfs. A reduction of 8.3% inthe total number of accidents was observed in thearea considered, while an increase of 47% wasrecorded in the region of control group.Selected assessments of traffic calming B/C-ratio

Road narrowing and road humps in residential areas inGermany 17.00


Source: FGSV (2001)

Speed humps on urban streets 2 to 4

Method: CBA Result: acceptable - excellent


Low cost traffic engineering measures in Greece 1.7 to 1.8



A before and after accident analysis with a large control group was used toevaluate the effects of low cost measures in Greece. The control group chosenconsisted of the neighbouring Municipalities of Holargos and Agia Paraskevi inthe Athens Greater Area.

53

2.4.7 Bypass roads

Over the years, existing road networks in urban areas can often not cope withexpected traffic volumes, even with extensive upgrading works. This results inincreased negative safety impacts for the affected communities and theircitizens, in particular vulnerable road users like cyclists and pedestrians. Abypass road on a new alignment promises to improve safety for regional trafficand local communities. Bypasses should mitigate adverse road safety effects ofthe transport network on local communities. Heavy traffic and other long-distance main road traffic can be moved away from local streets.

Bypasses are designed tocarry long-distance trafficoutside towns and cities, sothat conflicts between localtraffic and long-distancetraffic are avoided. Theconstruction of bypassesmakes it easier to introducetraffic calming measures onthe main road through atown, that when this roadserves through traffic.

Selected assessments of bypass roads B/C-ratio

Bypass roads in Norway and Sweden 0.84 –0.88



Bypass roads in Norway 1.03



Bypass roads increase mobility for both long-distance traffic and local traffic.Bypasses can make it easier for pedestrians and cyclists to cross roads intowns, since less traffic reduces waiting times. On the other hand, an increasein speed may make it more difficult to cross the road. A bypass road can be abarrier to local travel. The analysis of Elvik and Vaa (2004) include amongstothers reduced traffic volume on the old main road, reduced traffic noise,vibrations, local air pollution and estimates the costs of building a bypassaround NOK 20 million per km road.

54

2.4.8 Signal control at rural junctions

A road junction presents a natural point of potential conflict between differenttraffic streams. As traffic volumes increase, the probability of conflict increasestoo, and traffic delays worsen. Traffic signal control at intersections separatestraffic streams and improves the flow. Traffic signal control can be achieved byusing lights, which may be either time-controlled (phases change after a giventime irrespective of the amount of traffic) or vehicle-actuated (the length of thephases is adapted to the amount of vehicles up to a given maximum phaselength).

In Israel, some 10% of both injuryaccidents and fatalities occur atrural junctions (CBS, 2003). Whenthe accidents occur at unsignalisedintersections, the majority ofaccidents are usually right-angle,rear-end and pedestrian accidents.For unsignalised intersections,introducing traffic lights is frequentlysuggested as a safety treatment toreduce all accident types.

(Source: DVR)

Selected assessments of B/C-ratio

Signal control at rural junction in Israel 1.25



The CBA of the measure in Israel accounts for safety effects only; aconsideration of time savings would strengthen the benefits of the measure.International experience demonstrates (Elvik and Vaa, 2004) that the effect onaccidents of traffic signal control at intersections was mostly positive, providingon average a 15% accident reduction at T-junctions and a 30% accidentreduction at crossroads.

55

2.4.9 Roundabouts

The roundabout is a common form of intersection control used throughout theworld. An increasing number of intersections are being converted intoroundabouts. These have a traffic calming effect and help limit the severity ofany collision that might occur. Modern roundabouts can be used at a widevariety of intersections. An advantage of roundabouts is the reduced number ofconflict points compared with uncontrolled intersections. Decision-making issimple combined with a lower level of conflicts. Furthermore, roundabouts slowdown all vehicles. The tighter the curve the lower the speed, which means that itis easier to stop or at least a possible impact would be relatively minor.Consequently, roundabouts reduce the number and severity of accidents. Onthe other hand, especially in the beginning, roundabouts can be unfamiliar tothe average driver which could lead to more accidents. Nevertheless, crashesat roundabouts are primarily rear end or low speed merge crashes.

Example from Czech Republic: In 2003 more than 70% of accidents took placein urban areas and about 10% of them occured at four-arm intersections.Between 1995 and 2001, eight four-arm intersections in urban areas withouttraffic lights were rebuilt as four-arm roundabouts. An average accidentreduction of 37,6% was calculated.

Roundabout inside urban area (Source: DVR) Roundabout outside urban area (Source:ADAC)

Selected assessments of roundabouts B/C-ratio

Roundabouts in urban areas in Norway 1.23 – 8.61



Roundabouts in Norway and Sweden 1.52 - 2.26



Roundabout in urban areas in the Czech Republic 1.5



56

2.4.10 Roadside guard rails

Roadside guard rails can prevent collisions with solid obstacles such as treesalongside the roads, mostly outside towns. They are often set up along the sideof the road in places where circumstances render it impossible to make theroadside area safe. Under normal circumstances solid obstacles should have aminimum distance to the edge of the road, but if this is not possible for anyreason a guard rail should be installed at this place.

(Source: BASt)

Example from France: The RN 134,which crossed the forest of“Landes” (South West of France)along 64,5 km has long tree-linedstretches of road on which 38,5% ofthe accidents occurred againsttrees. The problem was to takemeasures to reduce the numberand the severity of the crashesalongside the stretch having thehighest and increasing level of risk(a stretch of 26,5 km). The measureconsists of the implementation of7800 meters of guardrails, 13frontage accesses and 8 lay-bys.The problem was here to proposeand negotiate measures to reducethe number and the severity ofcrashes by ensuring the protectionof the row of trees by the means ofguardrails when it is possible orotherwise by the mean of treefelling.

Selected assessments of Guard rails B/C-ratio

Measures against collisions with trees in France 8.69



Measures against collisions with solid obstacles onroads outside towns in Switzerland 32



Guard rails on the roadside in Norway and Sweden 0.69–1.18Method: CBA Result: poor - acceptable

Source: Elvik R. (1999), Elvik R. (2001), Elvik R. (2003), Elvik, R.; AmundsenA.H. (2000)

57

2.4.11 Middle guard rails and other lane barriers

Middle guard rails and other lane barriers are instaalled to prevent collisionswith oncoming vehicles on roads mostly outside towns, especially fatalcollisions caused by vehicles crossing the central line. Therefore, roads outsidetowns should be equipped with guard rails if the average daily traffic volume issufficiently high. Moreover, guardrails between foot and cycle paths and theroad can increase safety amongst pedestrians and cyclists using foot and cyclepaths.

Two sections of the Budapest Ring Road (M0 BRR) functions as a 2x2 laneexpressway. Traffic lanes are 3,50 m wide and there are no emergency lanes. Itwas decided to increase the physical central separation by placing additionalconcrete New Jersey elements. It has been demonstrated that the break-evenpoint is reached within less than 7 years. For a 10 years project life period, theB/C-ratio is around 1,35, reflecting strong economic efficiency of the measure.

New Jersey elements on the M0 Budapest Ring Road (Source: KTI)

Selected assessments of roadside and middle guard rails andother lane barriers B/C-ratio

Moveable lane barrier on a motorway bridge in NewZealand 6.80


Source: Leak, Hawkins, Sansom, Dunn (1993)

Placing additional New Jersey elements along the centralline of existing sections of M0 Budapest Ring Road (M0BRR) in Hungary

1.35


Source: Tomaschek, Tamás Attila (2002)

Installing bollards (plastic post delineators) at exit goreareas of highways in Israel 1.84 – 2.65


Source: Hakkert, Gitelman (1998)

58

2.4.12 Two plus one roads

Two plus one (2+1) road construction is a measure where an existing road isupdated to have a middle lane changing direction every 1 – 2.5 kilometre.Alternatively the construction method can be applied to new road sections, butsince the upgrading is a low-cost measure (compared for instance toconstruction of a new motorway), the standard application is to existing roadsections. In principle, the 2+1 road construction takes place on 13-meter-wideroads, and it is considered as means of upgrading other solutions, mainly wideshoulders or wide lanes. The distinctive advantage of the 2+1 solution is that itprevents the head-on collisions. A possible way to construct the 2+1 road is toset a fixed steel median cable on the road, which then shifts to the other sidewhen the overtaking lane is switched to the other direction.

(Source: BASt)

Head-on collisions have been a severe problem as a percentage of totalfatalities in Finland, between 1996-2000 an average of 80-85 percent of fatalaccidents on two-lane highways were due to head-on crashes.

Selected assessments of B/C-ratio

2+1 roads in Finland 1.25



2+1 roads in Sweden 2.26



The main source of benefits in the Swedish 2+1-roads was the safety impact(reductions in estimated deaths) and the time savings due to the overtakinglane.

59

2.4.13 Redevelopment of railway crossings

Even if the number of fatalities has decreased during the last decade, manypeople are still killed at level crossings. Too many drivers believe that they canstill cross even though traffic lights do not allow them to cross, andunderestimate the breaking distance of a train. Some of these accidents couldbe prevented by a better crossing design. Level crossings could be equippede.g. with or without signs, signals, half or full barriers. Especially open crossingsare considered to be more risky than crossings with gates. A grade-separationof a road-rail crossing means building a bridge or a tunnel instead of an existingat-grade crossing. Grade separation eliminates existing railway-road crossingsand consequently, removes the problem of train-vehicle collision at the siteconsidered It considerably diminishes road traffic delay at the site causedpreviously by crossing closures due to train movements. A grade-separatedcrossing is usually considered where crossings are already protected byautomatic gates.

The majority of rail-road crossings in Finland are level (at-grade) crossings. Inthe Finnish context, level crossings have been considered the cost-effectiveway to construct crossings, due to the fact that traffic volumes at crossings havebeen small. However, severe accidents at crossings in the late 1990s raisedawareness of the need to upgrade crossings, either through increased safetymeasures or by constructing grade-separated crossings.

(Source: DVR) (Source: ADAC)

Selected assessments of redevelopment of railwaycrossings B/C-ratio

Grade separation at road-rail crossings in Finland 0.25 to 0,65

Method: CBA Result: poor - acceptable


Grade separation at road-rail crossings in Israel 1.0 to 2.8



Grade-separation of an at-grade crossing can be beneficial under certainconditions. The daily number of trains and daily road traffic volume are the maincrossing parameters as they influence both the accident frequencies and theextent of traffic delays at the crossing.

60

2.4.14 Road lighting

At night, visual capabilities are impaired and visibility is reduced. In order todrive safely on roads at night drivers must be aware of the conditions aroundtheir vehicles and see other road users adequately far away from them. Roadlighting is a potential countermeasure to reduce the number of night-timeaccidents at locations with inadequate illumination. These locations, especiallyjunctions, intersections, access roads and tunnels, should be redevelopedregarding their road lighting equipment.

(Source: DVR)

Most of the information driversutilise in traffic is visual. Visualconditions can therefore be verysignificant for safe travel.

Many studies have shown that roadlighting reduces the number of fatalaccidents in the dark by nearly65%, the number of accidents in thedark involving personal injuries byalmost 30% and the number ofaccidents involving materialdamage by close to 15% (Elvik andVaa, 2004).

Selected assessments of road lighting B/C-ratio

Road lighting in Norway and Sweden 1.21 – 2.51



Upgrading substandard road lighting in Norway 2.62 - 4.32



Installing of road lights in Norway 7.23 – 9.25



61

2.4.15 Hazard warning

Safety, hazard and warning signs ensure that road users and emergency staffhave adequate information concerning specific dangers related to this roadsection. Hazard warning signs have to be consistent. A hazard warning shouldindicate a potentially hazardous situation which, if not avoided, could result indeath or serious injury. Confusion among the road users should be avoided byusing standardised patterns, phrases, colours, shapes and pictograms. The installation of utility poles with flashing lights could increase driversattention in dangerous road sections . The hazard warning sign should be usedbeside the road when it is needed to indicate the presence of a specific danger.

There are two types of hazard warning signs: those that warn of a permanenthazard, and those that warn of a temporary hazard.


Selected assessments of hazard warning B/C-ratio

Installation of utility poles with flashing lights in Switzerland 2.7



62

2.4.16 Preventing accidents with animals

Some road safety measures are dedicated especially to prevent accidents withanimals. A successful accident reduction technique requires a certainunderstanding of animal movement patterns and behaviour. One approach is toplace deer fences beside the roads, designed to keep these and other largeanimals from straying from the forests across the road. Another infrastructuralapproach is to build over- and underpasses for animals. Furthermore, odoroussubstances and reflective apparatus could be applied to prevent animals fromcrossing roads. Habitat alteration tries to solve the problem on an ecologicalbasis.


Selected assessments of preventing accidents with animals B/C-ratio

Preventing accidents involving animals in Norway andSweden 0.11 – 1.83

Method: CBA Result: poor -acceptable


Setting up stationary game danger signs has proved to be ineffective if used asthe sole measure to prevent accidents with game. The probability ofencountering game in these areas is too low for the road user to take the signseriously even if it is much higher than in other areas.

63

2.4.17 Measures regarding skid resistance

Road surface characteristics and conditions can influence the occurrence ofaccidents. All road surface structures gradually deteriorate with time. Thisdeterioration is normally evidenced by the appearance of various types ofsurface distress caused by a combination of environmental conditions and roaduse. If the road surface is not repaired, surface distress may become severeenough that road safety could be affected. Maintenance needs can be identifiedthrough pavement condition surveys. For example, skid resistance of thepavement surface must be maintained to provide for safe braking. Someaccidents happening on wet surfaces could be prevented if the road surfacemaintenance were carried out in time, assuming that wet weather crasheswould increase with lower skid resistance. Therefore, one might consider thatroad surfaces in risky conditions should be rehabilitated in accordance with afixed time schedule. When maintenance only is no longer effective, restorationshould be required.

(Source: DVR)

Selected assessments of road maintenance measures B/C-ratio

Management of skid resistance on the Australian roadnetwork 3.7 – 12.6


Source: Cairney (1997)

Skid resistance is the force developed when a tyre that is prevented fromrotating slides along the pavement surface. Many studies have shown thatweather crashes increase with lower skid resistance.

64

2.4.18 Winter road maintenance

Adverse weather conditions in winter can dramatically affect the roadtransportation system. In winter, with the road surface covered with snow or ice,the situation is almost always critical for road safety. Emergency brakingsituations occur frequently. Winter road maintenance should make roadspassable throughout winter times and thus can help to increase road safety ifwintry conditions occur. The intensity of winter maintenance on a road oftendepends on its traffic capacity and functional classification. To plough the roadnetwork clear of snow is one duty of winter road maintenance. Moreover,clearing winter roads to the bare pavement usually requires de-icing chemicals.Depending on the temperature and the desired result, materials for wintermaintenance like salt, calcium chloride, calcium magnesium acetate or sandhave to be chosen to ensure the effectiveness of maintenance operations.

A detailed plan has to be drawn up to make sure that roads are made clear ofsnow as quickly as possible according to their importance for the traffic flow.

(Source: DVR) (Source: ADAC)

Selected assessments of winter road maintenance B/C-ratio

Winter maintenance of roads in Norway and Sweden 2.67 – 3.17

Method: CBA Result: acceptable –excellent


Winter maintenance of roads increases the average speed of traffic and that iswhy it has a major effect on mobility. On the other hand several environmentalsideeffects are possible: e.g. increased salt content in soil and the groundwater.Salt also contributes to increased rust on cars but this effect is difficult to isolate.

65

2.4.19 Increasing construction site security

Construction firms could be obliged to provide a specific safety policy for theirconstruction sites to reduce the number of road accidents at the work zones .Road workzone related safety measures should be developed and applied tomitigate the adverse safety effects on workers and road users. Measures toreduce such accidents could include using more message signs to warnmotorists, scheduling work when traffic is lighter and putting specific barriersbetween workers and the cars and trucks driving past.

(Source: DVR)

In the Swiss proposal studied withinthe VESIPO program, constructionfirms have to provide a specificsafety concept for everyconstruction site. This concept mustbe approved by a safety audit.According to the concept, therealisation of the concept must beapproved by the responsibleaccident-commission. The mainpart of the costs for technical andorganizational safety measures atthe construction site have to beborne by the construction firmsthemselves.

Selected assessments of increasing construction site security B/C-ratio

Securing of highway construction sites in Switzerland 7.0



Many measures may increase safety at road works: e.g. temporary trafficcontrol with temporary speed limits, traffic signals, temporary road markings andflagging. Safety measures at road works are at least as important at night,especially as a result of reduced visibility and because the speed levels can behigher than during the day.

66

2.4.20 Accident prevention and performance monitoring: Centralisedmanagement of dangerous goods transport

If a truck carrying dangerous freight is involved in a road accident, an additionalrisk occurs if the freight escapes as a consequence of the accident. Centralisedmanagement of the transport of hazardous freight could reduce the danges. Itcould comprise procedures to permit, notify and monitor dangerous orhazardous freight transport, controlled by a centralised management unit.

In Switzerland the centralised management of dangerous or hazardous freighttransport includes a variety of procedures including: compulsory registration ofall dangerous freight transport; information and automatic assignment of themode, the time and the route of transport; the transport containers have to beequipped with devices to transmit automatically freight, vehicle and locationdata; automatic GPS tracing of all dangerous or hazardous freight transports;automatic control facilities to monitor all hazardous freight movements andtransport permissions; and centralised management division to preventcumulative risks and to observe risky freight transports locally and temporarily.

(Source: DVR)

Selected assessments of centralised management of dangerousgoods transport B/C-ratio

Centralised management of dangerous or hazardous freighttransports in Switzerland 0.11



Hazardous Goods Regulations in Norway 0,8


Source: Elvik, Vaa (2002)

Safety measures for hazardous goods may reduce the extent of discharge ofhazardous materials in accidents. This may limit damage to the environment.Quantified effects are currently not available.

67

References

Some comprehensive efficiency analyses of road safety measures have beencarried out e.g. in the USA, Switzerland and Scandinavian countries. For theUSA, the cost effectiveness of road safety measures was assessed by Tengs etal. for five hundred life saving interventions in 1995 (Tengs et al., 1995). Thecost effectiveness of life-saving interventions in the United States was analysedby performing a comprehensive search in publicly available economic analysesof life saving interventions (literature analysis). A life-saving intervention isdefined as any behavioural and/or technical strategy that reduces the probabilityof premature death among a specified target population. Cost-effectiveness isdefined as the net resource costs of an intervention per year of live saved (in1993 dollars). To improve the comparability of cost-effectiveness ratios arrivedat by diverse methods, fixed definitional goals are established, e.g.:

• Costs and effectiveness should be evaluated from the societal perspective(in contrast to, for example, the insured motorists` perspective).

• Costs and effectiveness should be "net" (any resource savings or mortalityrisks induced by the intervention should be subtracted out).

• Costs should be "direct" (indirect costs, such as foregone earnings, forexample, should be excluded).

• Future costs and live years saved should all be discounted to their presentvalue.

In the Swiss VESIPO(road traffic safety policy)-program, the economicassessment of road safety measures is one part of the whole study of“Definitions of the principles for a federal road traffic safety policy”. Within thiswork, the developed assessment model was used for 77 road safety measuresin total (VESIPO, 2002).The analysis is based on a model developed by the bfu Switzerland thatdeterminates the efficiency of safety measures. The costs are based on data ofthe statistical report 2001 of the Swiss Federal Bureau of Statistics, informationof experts plus additional internet and literature information. The calculation of the expected benefits is based on the monetized values forfatalities, injuries and property damages prevented by the measure:

• 1.92 Mio. CHF for every fatality.

• 247,000 CHF for every serious casualty.

• 9,000 CHF for every slightly injured crash victim.

• 28,000 CHF average property damage.

Recent analyses in Norway and Sweden (Elvik, 1999; Elvik, 2001; Elvik, 2003;Elvik and Amundsen, 2000) conclude that cost-effective road safety policiescould prevent between 50 and 60% of the current number of road accidentfatalities in both countries. In order to develop strategies for improving road

68

safety a broad range of road safety measures (132 for Norway, 139 forSweden) was screened for the purpose of determining potential effectiveness.

Abele, J.; Baum, H.; et al (2005)Exploratory Study on the potential socio-economic impact of the introductionof Intelligent Safety Systems in Road Vehicles, Germany, January 2005

Aberg, L. (1999)The role of attitudes in decisions to violate traffic regulations. Paperpresented at the European conference of transport psychology in Angers,France, June.

Amundsen, Astrid Helene; Rune Elvik and Lasse Fridstrøm (1999)Effects of the "Speak out!" campaign on the number of killed or injured roadusers in the county of Sogn og Fjordane, Norway. In: TØI report 425/1999

Anderson, G (2003a)Detection and avoidance of collisions: the ACT model.http://www.path.berkeley.edu/PATH/Research/AVCSS/vsafety2.html.

Andersson, G. (2003b)Automatisk hastighetsövervakning. Resultat från försöksverksamheten. VTIreport (draft), Swedish Road and Transport Research Institute. Linköping

Andreassen D.; Cusack S. (1996)Analysis of improving occupant protection in existing coaches. Report No MR13, Federal Office of Road Safety, Canberra, Australia.

BASt (2002)Sicherheitsaudit für Straßen (SAS) in Deutschland; in: Berichte der BASt;Verkehrstechnik Heft V98.

Baum, H., Höhnscheid, K.-J. (2001)Economic Evaluation of Road Traffic Safety Measures: Germany, in: ECMT,Round Table 117, Paris 2001; P. 5 – 40.

Begg, D., Langley, J. D., Stephenson, S. (2003)Identifying factors that predict persistent driving after drinking, unsafe drivingafter drinking, and driving after using cannabis among young adults. AccidentAnalysis and Prevention 35, 669–675.

Bliss, T., Guria, J., Vulcan, P., Cameron, M. (1998)The New Zealand Road Safety Advertising/ Policing Package and itsEvaluation. Road safety research, policing, education conference,Wellington, New Zealand, Vol. 2.

Bohlin,N (1967)A Statistical Analysis of 28000 Accident Cases with Emphasis on OccupantRestraint Value, Göteborg, AB Volvo.

Brilon, W. u.a. (2002)Wirkungsanalyse der Aktion "Darauf fahr ich ab...", Bochum, unpublished

Cairney, P. (1997)Skid resistance and crashes – a review of literature. Report ARR 311, ARRBTransport Research.

69

Cameron, M.H., Haworth, N., Oxley, J. et al (1993)Evaluation of Transport Accident Commission road safety televisionadvertising. Report No 52, Monash University Accident Research Centre,Victoria, Australia.

Carsten, O., Tate, F. (2000)External Vehicle Speed Control Project, Final Report: Integration. Universityof Leeds, Institute for Transport Studies.http://www.its.leeds.ac.uk/projects/evsc/contact.html.

Carsten, O.; Tate, F. (2001)Intelligent Speed Adaptation: the best collision avoidance system?Proceedings of the 17th International technical Conference on the EnhancedSaftey of Vehicles, 4-7 June, 2001, Amsterdam, the Netherlands.

Datta, Tappan K. (1998)Evaluation of low cost improvements at signalized Intersections.

Dravitzki V.K., Tate, F.N., Koorey, G.F., Davies, R.B. (1998)Effect of reflectorised number plates on “hit parked vehicles” in nighttime.Road safety research, policing, education conference, Wellington, NewZealand, Vol. 2, pp 114-119.

Eagly, A. H., Chaiken S. (1993)The psychology of attitudes. Fort Worth: Harcourt Brace Jovanovich.

Eckhardt, A., Seitz, E. (1998)Wirtschaftliche Bewertung von Sicherheitsmaßnahmen, bfu-Report 35, Bern1998

ECMT (2001)Economic evaluation of road traffic safety measures. Report of the 117 roundtable on transport economics; Paris, October 2000. European Conference ofMinisters of Transport.

Elvik, R. (1995)The safety value of guard rails and crash cushions: a meta-analysis ofevidence from evaluation studies. Accident Analysis and Prevention 27, no.4, 523–549.

Elvik, R. (1996)Does prior knowledge of safety effect help to predict how effective a measurewill be? Accident Analysis and Prevention 28, no. 3, 339–347.

Elvik R. (1999)Bedre trafikksikkerhet i Norge. Rapport 446. Transportøkonomisk institutt,Oslo

Elvik R. (2001a)Cost-benefit analysis of road safety measures: applicability andcontroversies. Accident Analysis and Prevention 33, 9-17.

Elvik R. (2001a)Improving road safety in Norway and Sweden: analysing the efficiency ofpolicy priorities. Traffic Engineering and Control, January 9-16.

Elvik, R. (2003)Assessing the validity of road safety evaluation studies by analysing causalchains. Accident Analysis and Prevention 35, 741–748.

70

Elvik, R., Borger-Mysen, A. and Vaa, T. (1997)Trafikksikkerhekshandbok (Traffic Safety Handbook). Institute of TransportEconomics, Oslo, Norway.

Elvik, R.; Amundsen A.H. (2000)Improving road safety in Sweden. Main report. Report 490. Institute ofTransport Economics, Oslo.

Elvik, R.; Vaa, T. (2004)The Handbook of Road Safety Measures; Amsterdam 2004.

Elvik, R.; Christensen, P. and Olsen, S.F. (2003)Daytime Running Lights. A Systematic Review of Effects on Road Safety.Report 688/2003. Institute of Transport Economics, Oslo, Norway.

Escape (2001)Working paper 1, "Cost-benefit analysis of police enforcement", March 2001.

European Commission (1994)HOPES Project, Guidelines on Safety Evaluation. DRIVE Project V2002,Bulletin 118.

European Commission–DG VII (1998)STAIRS Project, Final Report. http://www.vsrc.org.uk/stairs/index.htm.

Evans, A.W. (2001)Economic Evaluation of Road Traffic Safety Measures: United Kingdom, in:ECMT, Round Table 117, Paris 2001, S. 89

ETSC (2005)Alcohol interlocks, ETSC Fact Sheet Number 04, June 2005.

ETSC (2003)Cost effective EU Transport Safety Measures, Brussels 2003

FGSV (2001)Merkblatt für die Auswertung von Straßenverkehrsunfällen; Teil 2;Forschungsgesellschaft für Straßen- und Verkehrswesen; ArbeitsgruppeVerkehrsführung und Verkehrssicherheit.

Fildes B.N., Cameron, M.H., Vulcan, A.P. et al (1994)Airbag and facebag benefits and costs. Accident Analysis and Prevention,26(3), 339-346.

Fildes B.N., Digges K., Carr D., Dyte D., Vulcan A.P. (1995)Side impact regulation benefits. Report No CR 154, Federal Office for RoadSafety, Canberra, Australia.

Fildes, B., Langford, J., Deery, H. et al (1998)Licensing assessment for older drivers. Road safety research, policing,education conference, Wellington, New Zealand, Vol. 2, pp 198-203.

Fildes, B., Digges, K., Dyte, D. (1997)The benefits of the European frontal offset regulation for Australia. Journal ofTraffic Safety Medicine, 25 (1-2), pp3-6.

Fildes, B., Digges, K., Dyte, D., Gantzer, S., Seyer, K. (1996)Benefits of a frontal offset regulation. Repoprt CR 165, Federal Office ofRoad Safety, Australia.

71

Frindstom, L. (1998)TRULS: an econometric model of road use, accidents and their severity. 8thWCTR, Antwerp, July.

Gitelman, V., Hakkert, S., Doveh, E., Cohen, A. (2001)A study of safety effects of road infrastructure improvements under Israeliconditions. International Conference: Traffic Safety on Three Continents; VTIKonferens 18A, Swedish National Road and Transport Research Institute,Linkoping, Sweden (CD-ROM). Lalloo, R., Sheiham, A.,

Gorell, R. (2001)Monitoring local road safety schemes and MOLASSES. Proceedings of theGood Practice Conference, 20-22 June 2001, Bristol, UK.

Hadi, M. A., Aruldhas, J., Chow, L. F., Wattleworth, J. A. (1999)Estimating safety effects of cross-section design for various highway typesusing negative binomial regression. Transportation Research Record 1500,TRB, National research Council, Washington, D.C., 169–177.

Hakkert, A.S. and Gitelman, V. (1998)Consideration of Bollard Treatment at Exit Gore Areas. TransportationResearch Record 1635, Highway Safety Modeling, Analysis and Design,pp.133-139, Washington, DC.

Hakkert, A.S., Gitelman, V., et al (2001)Development of Method, Guidelines and Tools for Evaluating Safety Effectsof Road Infrastructure Improvements. Final report, T&M Company, Ministry ofTransport (in Hebrew).

Halcrow Fox. (1996)Accidents at signal controlled junctions and pelican crossings in Glasgow.The Scottish Office Central Research Unit.

Hauer, E. (1997)Observational before/after studies in road safety. Pergamon. U.S.A.

Hedman K. O.; Stenborg L. (1991)Traffic safety measures and Cost- Effectivness Ratios. Swedish TransportResearch Board & Swedish Road and Transport Research Institute.Linköping.

Hendrie, D., Lyle, G., Haley, J. (2001)Safety benefits of improvements in vehicle design since the introduction ofthe ANCAP crash test program. Proceedings of 17th Int. TechnicalConference on the Enhanced Safety of Vehicles. Amsterdam, theNetherlands, June 2001.

Herrstedt L. (1999)Road Safety Audit. Note 67, 1999. Road Directorate. Copenhagen.

Herrstedt L. (2000)Road Safety Audit. Note 74, 2000. Road Directorade. Copenhagen.

Holló Péter, Kajtár Katalin, Dr.Schváb János (2000)Az útmenti fasorok és a pályaelhagyásos balesetek egyes összefüggései[Some relationships between roadside alleys and run off-road accidents]Közúti és Mélyépítési Szemle, 2000. No 10. pp. 358-365.

72

Holló, P. (1999a)A közelmúlt közlekedésbiztonsági intézkedéseinek hatékonyságvizsgálata.[Efficiency analysis of road safety measures of recent past.] Közúti ésMélyépítési Szemle, XLIX. No.12, december 1999, pp. 475-481.

Holló, P. (1999b)Közúti közlekedésbiztonsági intézkedések hatékonyságvizsgálata, különöstekintettel a nemzetközi összehasonlítás néhány módszertani kérdésére.[Efficiency analysis of road safety measures with special regard to somemethodological problems in international comparison.] D.Sc. Thesis.Hungarian Academy of Sciences, Budapest,.

Hülsen, H.; Meewes, V. (1998)Economic Costs and Benefits of Countermeasures against RoadsideAccidents, 9th International Conference, Road Safety in Europe, September21-23 1998, Bergisch Gladbach/Cologne, Germany, Program 23/9.

Hylander D.; Elfstrand J. (2002)The Vision Zero Circuit in Trollhättan - a Cost-Benefit Analysis of theMunicipal Part. A 60 point essay in economics, 2002. University of KarlstadSweden..

IMMORTAL (2005)Detailed Cost-benefit analysis of potential inpairment countermeasures,

IMMORTAL Delivery D-P2, February 2005.Jourdan P.W. (1995)

Road safety audit: the Australian approach. Routes/Roads No 288 –September, pp. 77-85.

Kahane, Charles J. (1998)The long-term effectiveness of Center High Mounted Stop Lamps inpassenger cars and light trucks, NHTSA technical report number DOT HS808 696, http://www.nhtsa.dot.gov/cars/rules/regrev/evaluate/808696.html.

Kampen, Ir L.T.B. van (1994)Cost-benefit study concerning car front impact requirements to increase thecrash-safety of pedestrians and cyclists. SWOV report R-94-31,Leidschendam.

Kelly D.M., Blum Y. (1992)The development of a traffic surveillance and control system for the Ayalonfreeway in Israel. 6 Int. Conf. on road traffic monitoring and control, theInstitution of Electrical Engineers, London (IEE Conference publication 355),pp. 58-62.

Koornstra, M., Bijleveld, F. and Hagenzieker, M. (1997)The safety effects of daytime running lights. SWOV, the Netherlands.

Lähesmaa, J. (1997)Kotka-Hamina sääohjatun tien yhteiskuntataloudelli-suus. Tielaitoksenselvityksiä 36/1997. Helsinki. (In Finnish, with English abstract)

Leak, MJ, Hawkins, NV, Sansom, EP, Dunn, RCM (1993)Safety and cost effectiveness of the moveable lane barrier on the AucklandHarbour bridge. ITE 1993 Compendium of Technical Papers, pp 366-370.

http://www.nhtsa.dot.gov/cars/rules/regrev/evaluate/808696.html

73

Maher, M. J., Summersgill, I. (1996)A comprehensive methodology for the fitting of predictive accident models.Accident Analysis and Prevention 28, no. 3, 281–296.

McLean A.J., Fildes B.N., Kloeden C.N., Digges K.N. et al (1997)Prevention of head injuries to car occupants: An investigation of interiorpadding options. Report No CR 160, Federal Office of Road Safety,Canberra, Australia.

Midtland K. et al. (1995)Evaluation of the In-depth accident Investigations of the Swedish NationalRoad Administration. TÖI report 296/1995. Oslo.

Nazroo, J. Y. (2003)Behavioural characteristics and accidents: findings from the Health Surveyfor England, 1997. Accident Analysis and Prevention 35, 661–667.

NET Engineering (2000)Piano Provinciale della Viabilità di Padova,

Newstead S.V. and Corben B.F. (2001)Evaluation of the 1992-1996 Transport Accident Commission fundedaccident black spot treatment program in Victoria. Report 2001/05 (182),Monash University, Victoria, Australia.

Newstead, S.V, Cameron, M.H., Leggett, L.M.W. (2001)The crash reduction effectiveness of a network-wide traffic police deploymentsystem. Accident analysis and Prevention, 33, pp 393-406.

Nilsson, G.; Andersson, G.; Björketun, U.; Larsson, J. (2002)Några trafiksäkerhetsåtgärder och samhällsekonomi. VTI notat 71-2001. VTI.Linköping..

Nokkala, M.; A. Schirokoff (2001)The socio-economic profitability of weather controlled road. Case Kotka-Pyhtää. (In Finnish, with English abstract). Tielaitoksen julkaisuja 51/2001. 52pages. Helsinki .

Norinder A. L.; Svensson M. (2003)Cykelhjälmskampanjer i Göteborg - en utvärdering av lönsamheten avsatsade medel. IHE arbetsrapport 2002:1. The Swedish Institute for HealthEconomics. Lund.

Ogden, K.W. (1996)Safer Roads: A Guide to Road Safety Engineering. Avebury Technical,England.

Ogden, K.W. (1997)The effects of paved shoulders on accidents on rural highways. AccidentAnalysis and Prevention, 29(3), 353-362.

ROSEBUD WP4 (2005)Testing the efficiency assessment on selected road safety measures, 2005

Promising (2001)Promotion of Measures for Vulnerable Road Users, Deliverable D5 "Cost-benefit analysis of measures for vulnerable road users, July 2001.

74

RTA (1995)Accident investigation and prevention. Procedures for road basedcountermeasures. Roads and Traffic Authority, New South Wales.

Saelensminde K. (2002)Walking and cycling track in Norweigan cities. Cost-benefit analysesincluding health effects and external costs of road traffic. TÖI report567/2002. Oslo.

Schlegel,H. (1972)Wirtschaftlichkeitsanalyse von Sicherheitskomponenten imPersonenwagenbau, Automobiltechnische Zeitschrift, 74.v., No.12, 1972.

Statens Vegvesen & Hamar kommune (1993)Trafikksikring i Hamar.

Schwarze, R et al (2005)

Is the market classification of risk always efficient?, Evidence from Germanthird party motor insurance, in: London School of Economics and PoliticalSciences, discussion paper no 32, February 2005. Also combined with H.J.Heinzmann and F.-D. Schade (2004); Risk groups in penalty point register -a base for fair prices in third party motor insurance; H-J. Ewers , Chr.Growitsch, R. Schwarze, Th. Wein, and H.P. Schwintowski (2004): Risk-orientated pricing in third party motor insurance(both reports published inGerman language, only)

SWOV (2002)Retro-reflecting contour marking on lorries: effective measure, limited effect.Research Activities, 21, November 2002.

Tengs, Tammy O; Adams, Miriam E.; Pliskin, Joseph S.; Safran, Dana Gelb;Siegel, Joanna E.; Weinstein, Milton C.; Graham, John D. (1995)Five–hundred live saving interventions and their cost-effectiveness, RiskAnalysis, Vol. 15, No. 3.

Tomaschek, Tamás Attila (2002)Az elválasztó betonelemek kihelyezésének és üzemeltetésének tapasztalataiaz M0 autóúton. Közúti és Mélyépítési Szemle, 2002/10, pp. 413-414.

Traven, A., Maraste, P., Persson, U. (2002)International comparison of costs of a fatal casuality of road accidents in1999; In: Accident Analysis and Prevention, 2002, S.323-332

Trivector Information AB (2002)Själv Säker. Reports 1999:5, 2000:7, 2002:2, Själv Säker 2002 (draft).Trivector. Lund

Turner, S., Nicholson, A. (1998)Intersection accident estimation: the role of intersection location and non-collision flows. Accident Analysis and Prevention 30, no. 4, 505–517.

Vagverket–Swedish National Road Administration (2003)Results of the world’s largest ISA trial. http://www.isa.vv.se/index.en.htm.

VESIPO (2002)Erarbeitung der Grundlagen für die Strassenverkehrssicherheitspolitik desBundes. u.a.; Basler & Hofman, Eckhardt, Perrin, Schönenberger, Fierz.

75

Walz, Marie C. (2001)The effectiveness of head restraints in light trucks,

Wesemann, P. (2001)

In: ECMT (2001). Economic evaluation of road traffic safety measures.Report of the 117 round table on transport economics; Paris, October 2000.European Conference of Ministers of Transport; P. 41 - 76.

thematic network...1 thematic network road safety and environmental benefit-cost and...

Documents