An evolutionary algorithm for the open vehicle routing problem with time windows

Abstract

This chapter presents an Evolutionary Algorithm (EA) for solving the Open Vehicle Routing Problem with Time Windows (OVRPTW). The OVRPTW is a well known hard combinatorial optimization problem that seeks to design a set of non-depot returning vehicle routes in order to service a set of customers with known demands. Each customer is serviced only once by exactly one vehicle, within fixed time intervals that represent the earliest and latest times during the day that service can take place. The objective is to minimize the fleet size following routes of minimum distance. The proposed EA manipulates a population of μ individual via an (μ + λ)-evolution strategy. At each generation, a new intermediate population of λ offspring is produced via mutation based on arcs extracted from parent individuals. The selection and combination of arcs is dictated by a vector of strategy parameters. The values of these parameters depend on the appearance frequency of each arc within the population and the current diversity of the latter, while their self-adaptation is facilitated utilizing a multi-parent recombination operator. Finally, each new offspring is further improved via a short-term memory Tabu Search algorithm and a deterministic scheme is followed for the selection of survivors. Experimental results on benchmark data sets of the literature illustrate the efficiency and effectiveness of the proposed bio-inspired solution approach. © 2009 Springer-Verlag Berlin Heidelberg.