General metaheuristic algorithm for a set of rich vehicle routing problems

Abstract

This paper studies a set of rich vehicle routing problems incorporating various complexities found in real-life applications. The rich vehicle routing problem considers simultaneously four multiple constraints: multiple depots, multiple time windows, multiple trips, and multiple vehicle types. A metaheuristic algorithm called the general vehicle routing algorithm, based on the skewed variable neighborhood search, was designed to address the problem of any combination on fve features. A slice of operators and heuristic approaches developed for the specifc constraints is embedded in the general vehicle routing algorithm. Six combination vehicle routing problem types were investigated. The computational results demonstrated that the proposed algorithm is competitive for both benchmark instances and generated instances in accuracy of solution and computational time.