Hybrid large neighbourhood search algorithm for capacitated vehicle routing problem

• Published in: Expert systems with applications Vol. 61; pp. 28 - 38
• Main Author: Akpinar, Sener
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2016
• Subjects: Algorithms; Ant colony optimization; Capacitated vehicle routing problem; Computation; Construction; Expert systems; Hybrid meta-heuristic; Large neighbourhood search; Mathematical models; Route selection; Search algorithms; Ant colony optimization; Capacitated vehicle routing problem; Large neighbourhood search; Hybrid meta-heuristic
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2016.05.023

•We aim at developing a new hybrid large neighbourhood search algorithm for CVRP.•Our algorithm incorporates solution construction heuristic of ACO into LNS.•The proposed hybrid LNS-ACO algorithm is tested on a set of CVRP instances.•Computational results indicate the satisfactory performance of the algorithm. This paper presents a new hybrid algorithm that executes large neighbourhood search algorithm in combination with the solution construction mechanism of the ant colony optimization algorithm (LNS–ACO) for the capacitated vehicle routing problem (CVRP). The proposed hybrid LNS–ACO algorithm aims at enhancing the performance of the large neighbourhood search algorithm by providing a satisfactory level of diversification via the solution construction mechanism of the ant colony optimization algorithm. Therefore, LNS–ACO algorithm combines its solution improvement mechanism with a solution construction mechanism. The performance of the proposed algorithm is tested on a set of CVRP instances. The hybrid LNS–ACO algorithm is compared against two other LNS variants and some of the formerly developed methods in terms of solution quality. Computational results indicate that the proposed hybrid LNS–ACO algorithm has a satisfactory performance in solving CVRP instances.