Integration of information and optimization models for vehicle routing in urban areas

Abstract

This paper is about efficient and reliable vehicle routing in urban areas. We enforce customer-oriented tour generation by provision and integration of time-dependent information models. Information models represent typical states of the traffic network in terms of time-dependent travel time data sets. In recent years, large amounts of telematics based traffic data have become available. Common optimization models are not capable of processing such data. Thus, we refer to a Data Mining approach that generates time-dependent information models by sophisticated filtering and aggregation of telematics based traffic data. The main focus is on the integration of different types of time-dependent information models into optimization models. Several information models and time-dependent vehicle routing heuristics are introduced and compared regarding data processing, computational effort and their impact on efficiency and reliability of pickup and delivery tours in urban areas.