This paper introduces a road safety analysis for different age and gender categories of road users. In contrast with many previous studies, time series of road crashes per age and gender category will be considered. The objective of the paper is to analyze Belgian data on the yearly number of fatalities per age and gender group, using a decomposition of the number of fatalities in terms of exposure and risk, in a time series perspective. For each category, a state space time series model is developed for the risk, which is defined as the number of fatalities divided by the magnitude of the population. It was found that road risk is changing over the age groups according to a U-shaped curve, and that men generally have a higher risk than women. Further, the risk is decreasing over time, but not at the same rate for all age-gender groups. The highest yearly reduction in risk is found for the oldest and youngest road users. The models are also useful to assess the attainability of formulated road safety targets, which makes them useful policy instruments. Especially for young males, the reduction in risk is not in line with Belgian and Flemish policy expectations. Van den Bossche, Wets and Brijs 3 INTRODUCTION This paper introduces a road safety analysis for different age and gender categories of road users. In contrast with many previous studies, time series of road crashes per age and gender category will be analyzed. The objective is to enhance the insights in the different characteristics of age and gender groups in terms of their crash involvement, thereby exploring the possibilities of time series analysis by means of state space models. The approach followed in the present study can be summarized as follows. First, the analysis focuses on road safety for road users of a specific age and gender category. It is often found in literature that road risk varies with age and gender. According to Evans [1], the number of fatalities among drivers in the US shows a peak at the age of 19 years old. The number of fatalities then steadily decreases with age. The completely different behavior in traffic of the age groups will probably be reflected in their crash records. The models developed here are therefore “disaggregated” from a road safety point of view. They describe the road safety situation for various subgroups of road users (per age and gender) and are therefore called “descriptive subset models”. Second, the number of fatalities will be described in two dimensions, namely the level of exposure and the level of risk. The first dimension, the level of exposure, describes the magnitude of the activity that results in fatalities, and is usually measured in terms of the number of trips, the number of vehicle kilometers, the trip duration or, in the absence of these data, a proxy like the level of the population, as is the case in the study at hand. It accounts for the number of potentially dangerous situations, or the exposure to risk. The second dimension is the probability of a fatality or the risk, given a certain level of exposure. Changes in one of these dimensions will change the safety situation. The two dimensions are naturally related to one another in a multiplicative way: ( ) ( ) ( ) Risk Exposure Exposure Fatalities Exposure Fatalities × =
[1]
S Lassarre,et al.
Analysis of progress in road safety in ten European countries.
,
2001,
Accident; analysis and prevention.
[2]
Gwilym M. Jenkins,et al.
Time series analysis, forecasting and control
,
1971
.
[3]
Siem Jan Koopman,et al.
Stamp 5.0 : structural time series analyser, modeller and predictor
,
1996
.
[4]
P. Vignal.
[Traffic safety].
,
1952,
Gazette medicale de France.
[5]
S Oppe,et al.
The development of traffic and traffic safety in six developed countries.
,
1991,
Accident; analysis and prevention.
[6]
Steven C. Wheelwright,et al.
Forecasting methods and applications.
,
1979
.
[7]
Andrew Harvey,et al.
Forecasting, Structural Time Series Models and the Kalman Filter.
,
1991
.