This paper illustrates a methodology developed to analyze the influence of traffic conditions, i.e. volume and composition on accidents on different types of interurban roads in Spain, by applying negative binomial models. The annual average daily traffic was identified as the most important variable, followed by the percentage of heavy goods vehicles, and different covariate patterns were found for each road type. The analysis of hypothetical scenarios of the reduction of heavy goods vehicles in two of the mostrepresentative freight transportation corridors, combined with hypothesesof total daily traffic mean intensity variation, produced by the existence or absence of induced traffic gives rise to several scenarios. In all cases a reduction in the total number of accidents would occur as a result of the drop in the number of heavy goods transport vehicles, However the higher traffic intensity, resulting of the induction of other vehicular traffic, reduces the effects on the number of accidents on single carriageway road segments compared with high capacity roads, due to the increase in exposure. This type of analysis provides objective elements for evaluating policies that encourage modal shifts and road safety enhancements. (A) Reprinted with permission from Elsevier.
Abstract