Integrating traffic safety in vehicle routing solution

Abstract

Traditionally, travel time was considered as the main travel cost in Vehicle Routing Solutions. This paper aims to develop a new travel cost function in the route selection process that integrates the crash risk cost estimation with the travel time cost in a comprehensive routing system. The proposed methodology estimates the cost of both the delay time and crash risk per road segments and intersections in the roadway network. The crash risk was predicted by applying Negative Binomial Regression model with Empirical Bayes function for fatal and injury crashes, separately. Two values of travel cost were estimated one based on the travel time and the other based on the crash risk probability. Five alternatives of combined sharing ratio between the two cost values are investigated to find a single cost value. A case study area was selected from Abu Dhabi roadway network. The optimum route has been decided based on network analyst module in Arc-GIS software package. The results showed that the integrated traffic safety element has an impact on the route selection where at many cases the optimum route has been changed based on the sharing percentage of the crash risk cost in the total journey cost.