A hybrid heuristic approach for the multi-commodity one-to-one pickup-and-delivery traveling salesman problem
- 371 Downloads
This paper addresses an extension of the Traveling Salesman Problem where a vehicle with a limited capacity must transport commodities. Each commodity has a weight, and exactly one origin and one destination. The objective is to find a minimum length Hamiltonian tour satisfying all the transportation requests without ever violating the capacity constraint. We propose for this problem a hybrid heuristic approach that combines the GRASP and VND metaheuristic techniques. Two variants of the method are presented, one of them using a mathematical programming based local search. We conduct computational experiments to compare the developed algorithms. The experiments show that they improve the best known solutions for a set of instances from the literature, and are able to cope with instances with up to 300 customers and 600 commodities in a reasonable amount of computation time.
KeywordsPickup and delivery Traveling salesman problem Hybrid heuristics
This work has been partially supported by “Ministerio de Ciencia e Innovación”, Spain (research project MTM2009-14039-C06-01).
- Ascheuer, N.: Hamiltonian path problems in the on-line optimization of flexible manufacturing systems. Ph.D. Thesis, Technische Universität, Berlin, Germany (1995) Google Scholar
- Maniezzo, V., Stützle, T., Voß, S. (eds.): Matheuristics. Hybridizing Metaheuristics and Mathematical Programming. Annals of Information Systems. Springer, Berlin (2009) Google Scholar