Fred Wegman final.ppt - International Transport ForumFred Wegman SWOV Institute for Road Safety Research Th N th l dThe Netherlands Fred Wegman September 2008 the good, the bad and

Fred WegmanSeptember 2008

the good, the bad and the potential for improvementwww.swov.nl

The Good, the Bad and … the potential for improvement

Fred WegmanSWOV Institute for Road Safety Research

Th N th l dThe Netherlands



Fatalities in main OECD/ITF regions(OECD/ITF to be published)(OECD/ITF, to be published)



Development in OECD/ITF regions

200EU15: 91 00: 3 1% / 00 07: 5 4%European Union (15)

150

EU15: 91-00:-3,1% / 00-07:-5,4%

NMS12: 91-00:-2,3% / 00-07:-0,8%

Europe-rest: 91-00:-1,7% / 00-07:-5,04%

Asia: 91 00: 3 5% / 00 07: 5 6%

European Union (15)EU New Member States (12)Other European CountriesOECD Asia Pacific

100

Asia: 91-00:-3,5% / 00-07:-5,6%

N-am: 91-00:0,2% / 00-07:0,2%

OECD Asia PacificNorth America

50

0

Traffic fatalities

1991=100



01970 1975 1980 1985 1990 1995 2000 2005 2010

“PRESS RELEASE“PRESS RELEASE 15th September 2008 Progress in road safety slowing down g y gRoad fatalities for the countries for which data are available show a slowdown in the downward trend of recent years (see Table 1 below) It is importantrecent years (see Table 1 below). It is important, however, to consider the data within a larger timeframe…..”



Predicted road traffic fatalities(World Bank Kopits/Cropper 2003)(World Bank, Kopits/Cropper, 2003)

250%

300%South Asia

East Asia &

150%

200%PacificMiddle East &North AfricaLatin America &C ibb

50%

100%CaribbeanEurope & CentralAsiaGlobal Total

0%2000 2020

High-income



Evolution in fatality and injury crashes(OECD/ITF to be published)(OECD/ITF, to be published)



We have more than fatalities ….

• A less positive conclusion can be drawn about progress made in OECD/ITF countries when using injury figures than figures on fatalities

• It is recommended to add injury data to internationalIt is recommended to add injury data to international databases (such as IRTAD) based on an international agreement on definitions and on how to

d i h i d d irespond in a harmonized way to underreporting



Progress of mortality rates by age (I)

Rate by million population, EU144001991

300

19982006

200

100

0



0 20 40 60 80 100

Progress of mortality rates by age (II)

Rate by million population, 20% y p pchanges EU14

0%0 20 40 60 80 100

-40%

-20%

-60%

-40%

-80% 1991-19981998-20061991 2006



-100% 1991-2006

Progress by mode of transport

140

120

1996

=100

100

y m

ode,

inde

x

80

fic fa

talit

ies

by

Pedestrian -5%

Bicycle -4,2%

Moped -5,2%

40

60

traf

f Motorcycle 0,5%

Car driver -2,8%

Car passenger -4,5%

Other -2,5%



1990 1995 2000 2005 2010

Motorcyclists EU-15 per age group

6

7

8

9

12 - 17 years250

300

350

18 - 24 years1200

1400

1600

1800

2000

0 - 11 years0

1

2

3

4

5

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

50

100

150

200

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

200

400

600

800

1000

1200

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012

400

500

600

700

800

900

1000

800

1000

1200

1400

1600

1800

40 - 49 years600

800

1000

1200

25 - 29 years0

100

200

300

400

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012

30 - 39 years0

200

400

600

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

200

400

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012

50 59600

60 74140

160

7545

50

50 - 59 years

100

200

300

400

500 60 - 74 years

40

60

80

100

120

140 75+ years

10

15

20

25

30

35

40

45



0

100

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

20

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

5

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012

Relative fatality rate and its relative annual reduction for different settlement types of Belgian communes

1,201,30

1,401,50

0,901,001,10

1,20

0 500,60

0,700,80

0,50

C ity centre Inner c ity Inner s uburbs Rural Outer s uburbs

R is k Trend



Mortality for different settlement types;EU 25 countries in 2004 (Eksler)EU-25 countries in 2004 (Eksler)

2,5

2

2,5

1

1,5

0

0,5

0Very densely

pop. withlarge centres

Densely pop.with large

centres

Densely pop.with large

centres

Densely pop.without large

centres

Less denselypop. withcentres

Less denselypop. without

centres



Regional differences for mortality and population density in the Netherlandspopulation density in the Netherlands

12

REOv

Dr

Zl

GldFr

10

12

habi

tant

s

RAN

LbNb

Gr

RHRA

Ut

ZhNh

RT

GldFl

6

8

100,

000

inh

RHRRRU

Zh

2

4

talit

ies

per

00 500 1000 1500 2000 2500

Inhabitants per square kilometer

Fat



ab ta ts pe squa e o ete

Mortality rates for NUTS-2 regions in EU (2004)in EU (2004)

1,1 - 5,4

5,5 - 8,9

NUTS 2Canaries (ES)

9,0 - 12,8

12,9 - 18,6

18,7 - 32,4

Açores (PT)

Madère



Fond : RRG GIS database

Evolution road fatalities in the EU



Progress on fatalities in the EU

80.000European Union 27 Member States

60.000

European Union 15 Member States

European Union New Member States

40.000

20.000

All trends on 1991-2000

0



01990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012

All countries move to the same spot ?!Fatality rate vs mortality rateFatality rate vs. mortality rate

fatality rate vs . mortality rate for 20 E uropean c ountries

0,09

0,07

0,08Aus tria

B elgium

C zech Rep.

Denmark

F inland

0,05

0,06

fatalities

/population) F inland

F rance

Germany

Great B ritain

G reece

Hungary

0 02

0,03

0,04

mortality rate (f g y

Ireland

Italy

Luxemburg

Netherlands

Norway

0

0,01

0,02

fa ta lity ra te (fa ta lities/number of vehic les)

P ortugal

S pain

S weden

S witz erland

P oland



0 0,5 1 1,5 2 2,5

An example: number of traffic fatalities in the Netherlands (running total)

11661200

in the Netherlands (running total)

One third of 20% drop could be explained by

1085

1001

1033

10831069

1088

1081

10281000

1100

s

could be explained by speed, alcohol, seatbelt and less moped use

996 1001

881

993 987

1028

836

900

1000

fic fa

talit

ies

817 811791

804

730

836

760 770754

750

700

800Traf

f

l NL d 709

600

700

00 01 02 03 04 05 06 07 08 Year

real NL-roadsInhabitantsReported



200

200

200

200

200

200

200

200

200 Year

How to speed up our learning curve?

• We have to learn more from ex-post evaluations• Not only from high-impact, short-term and more or

less isolated interventions; progress is coming from many small steps forward in an ever changing worldmany, small steps forward in an ever changing world

• We have to improve our ex-ante evaluations to support decision making on road safety programmes

• Scientific Research on Road Safety ManagementWorkshop in the Netherlands 2009Special Issue Safety Science 2010



How to measure progress? Example from New ZealandExample from New Zealand

Social cost

Deaths

HospitalisationsHospitalisations

Drunk drivers killed

Open road speed

Urban speedUrban speed

Seatbelts not worn

Peds, cyclists killed + hosp.



2001 baseline 2004 target

Road Safety Benchmarking

• Not only fatality rates and mortality rates• The process of measuring various aspects of a road

safety performance of a country (or other jurisdiction) and comparing this with the performance of othersand comparing this with the performance of others, i.e. the best-of-class by identifying, understanding and adapting their (outstanding) practices

Who performs well?Who is the most compatible to benchmark with?What can I learn?



A framework for our knowledge:road safety target hierarchy (SUNflower)road safety target hierarchy (SUNflower)

Social costs

Number killed and injuredOutcome

Safety performance indicators

P li fSafety measures and programmes

Structure and culture

Policy performance

Policy context



Composite indicator for benchmarking purposesbenchmarking purposes

• Three entrances:Outcome indicators (final and intermediate outcomes)Quality of road safety ‘measures and programmes’Indicators on ‘structure and culture’Indicators on structure and culture

• SUNflower in SafetyNet (initial results later this year)



Towards a composite indicator forROAD SAFETY PERFORMANCE ?ROAD SAFETY PERFORMANCE ?

• AdvantagesSimplificationQuantificationCommunicationCommunication

• Accepted in many other fields, e.g.Financial world: Dow Jones, CAC, Human Development IndexEnvironmental Sustainability IndexyOverall Health System Index



Go fishing where fish is, but ….

• Look for high risks, high proportions, high increasese.g. novice drivers, elderly road users, PTW, high-risk locations

• Road crashes can occur and will occur everywhereRoad crashes can occur and will occur everywhere• We were (relatively) successful in fishing where the

fishes are• However, fishes are more and more everywhere• The answer is a systems approach



Road fatalities are scattered



Our fundamental road safety problem

• Today’s road traffic is inherently unsafe• Today’s road traffic is inherently unsafe• The road system of today has not been designed

with safety in mind, as is the case with air transport y , por rail transport

• Which means we are almost fully dependent on h th d k i t k iwhether a road user makes a mistake or error in

preventing a crash• Another approach is needed: Safe Safety Approachot e app oac s eeded Sa e Sa ety pp oac



To conclude

• We are all Good and Bad• All countries/regions have potential for improvements • Road Safety Management could be improved considerably• Which approach? Safe System ApproachWhich approach? Safe System Approach



Fred Wegman final.ppt - International Transport ForumFred Wegman SWOV Institute for Road Safety Research Th N th l dThe Netherlands Fred Wegman September 2008 the good, the bad and

Documents