Over the last decennia, the number of kilometres travelled on roads has been steadily growing. Mobility has become part of our life. This is a favourable thing, since the more people can move, the more they can participate in all sorts of activities. But on the other hand, traffic also has negative consequences. As long as people travel along public roads, there have been road accidents and victims. Traffic accidents restrict people in their social activities, and they limit other road users in their mobility. It is obvious that road safety and mobility are closely related. The basic pieces of information needed to investigate this relation are the number of victims and a measure of exposure. More specifically, the number of victims can be decomposed in terms of exposure and risk. The advantage of this approach is that the number of victims can be seen as the upshot of two forces working at the same time. A fatality results from the exposure to the risk and the probability of a fatality, given a certain level of exposure. Even if the risk has decreased over time, the number of fatalities may still be high because of an increased exposure. However, the main difficulty when studying this relation is the availability of appropriate data. The available exposure data is limited in several ways: the data quality is low, some measures of exposure are simply not available and if they are, they are rarely in a useful format. Clearly road safety is generally not the primary objective when gathering travel data. However, if we want to study the relation between road safety and mobility, we must make shift with what we have. This is the main reason for writing this paper. Given the available data in Belgium and in Flanders, we investigate the possibilities to study the relationship between mobility and road safety. In this study, we introduce some relatively simple models to enhance the insight in the relation between road safety and mobility. More specifically, we propose two ways of combining mobility and traffic safety data. In a first study, we use the yearly number of vehicle-kilometres and fatalities for Belgium (1973 - 2001) to decompose the road safety problem in risk and exposure. The main objective of these models is to monitor the evolutions in traffic safety. For policy makers, it is important to explain the trends in road safety. The models allow predicting the trends in the number of fatalities, giving an indication of policy effectiveness. Given the ambitious road safety targets set on a European level for 2010, this model is an instrument to assess the effectiveness of ongoing road safety programs. Next we use Flemish travel survey and accident data for the year 2000 to study disaggregated factors of mobility that determine traffic safety. It is known that different user groups (in terms of transport mode, age and gender) can have different patterns of traffic and a corresponding level of risk and exposure. When sustainable transport modes are promoted, policy makers have to make sure that these are safe and useful for the target group. It is important to find out whether a transport means is unsafe for a given user group because of a high level of exposure, or rather because of a higher level of risk. The results of this kind of analysis can put the relation between road safety and mobility in a different light. The classical decomposition of yearly data enhances the insight in the development of risk and exposure over time, and in how they determine road safety. This approach is also useful to predict the trend in road safety, which allows benchmarking the ongoing safety efforts. The second study extends the classical approach to a decomposition of road safety by age and gender. For each age-gender combination, the number of victims can be explained as the result of exposure to risk and the risk itself. The models include socio-demographic characteristics like age and gender and highlight the road users at risk. They are also useful to indicate the safety differences in the various modes of transport. The models that will be developed in this study should lead to an improvement of the analysis of factors that influence the accident process. For policy makers, this is key information for decision making, as it allows tackling the road safety problem at the source, which is a necessity for a sustainable transport policy. The models can steer road safety campaigns, and determine what kind of transport mode should be made more safe and more attractive for specific groups of road users. For the covering abstract please see ITRD E135207.
Abstract