A short note on applying a simple LS/LNS-based metaheuristic to the rollon–rolloff vehicle routing problem

Abstract

The rollon–rolloff vehicle routing problem (RRVRP) arises when tractors move containers or trailers between locations generating a high volume of waste like construction sites, etc., and a disposal facility. It can be formulated as a node routing problem with asymmetric arc cost and a maximum route length. In this paper we show that a simple and flexible approach which combines standard local search and large neighborhood search moves under two parameter-free/-poor metaheuristic controls outperforms special purpose developments published in the literature. Our results are obtained by customizing a RRVRP-solver from a framework developed for solving rich vehicle routing problem variants.