A modified ALNS algorithm for vehicle routing problems with time windows

Abstract

In this paper, we present an efficient heuristic for the vehicle routing problem with time windows (VRPTW), inspired by the Adaptive Large Neighborhood Search (ALNS) previously suggested by R pke and Pisinger [21]. The proposed heuristic uses the Modified Choice function (MCF) of Drake [12] as an elegant selection mechanism to favor the most successful operators instead of the roulette wheel selection. This general method is denoted Modified Adaptive Large Neighborhood Search (MALNS). The computational experiments are performed and the comparison with the classical ALNS is given according to Solomon's benchmark, and its extension the instances of Gehring and Homberger's benchmark.