Journal of Heuristics

, Volume 10, Issue 5, pp 483–506

A New Memetic Algorithm for the Asymmetric Traveling Salesman Problem

  • Luciana Buriol
  • Paulo M. França
  • Pablo Moscato
Article

DOI: 10.1023/B:HEUR.0000045321.59202.52

Cite this article as:
Buriol, L., França, P.M. & Moscato, P. Journal of Heuristics (2004) 10: 483. doi:10.1023/B:HEUR.0000045321.59202.52

Abstract

This paper introduces a new memetic algorithm specialized for the asymmetric instances of the traveling salesman problem (ATSP). The method incorporates a new local search engine and many other features that contribute to its effectiveness, such as: (i) the topological organization of the population as a complete ternary tree with thirteen nodes; (ii) the hierarchical organization of the population in overlapping clusters leading to the special selection scheme; (iii) efficient data structures. Computational experiments are conducted on all ATSP instances available in the TSPLIB, and on a set of larger asymmetric instances with known optimal solutions. The comparisons show that the results obtained by our method compare favorably with those obtained by several other algorithms recently proposed for the ATSP.

asymmetric traveling salesman problem local search memetic algorithms metaheuristics 

Copyright information

© Kluwer Academic Publishers 2004

Authors and Affiliations

  • Luciana Buriol
    • 1
  • Paulo M. França
    • 1
  • Pablo Moscato
    • 2
  1. 1.Faculdade de Engenharia Elétrica e de Computação, Universidade Estadual de CampinasUNICAMPCampinas, SPBrazil
  2. 2.School of Electrical Engineering and Computer Science, Faculty of Engineering and Built EnvironmentThe University of NewcastleCallaghan, NSWAustralia

Personalised recommendations