Performance analysis of the multi-objective ant colony optimization algorithms for the traveling salesman problem

Abstract

Attached

Most real w orld combinatorial optim ization problem s are difficult to solve w ith m ultiple objectives w hich have to be optimized simultaneously. Over the last few years, researches have been proposed several an t colony optim ization algorithm s to solve m ultiple objectives. The aim of this paper is to review the recently proposed multi-objective an t colony optim ization (MOACO) algorithm s and compare their perform ances on two, three and four objectives w ith different num bers of ants and num bers of iterations. Moreover, a detailed analysis is perform ed for these MOACO algorithm s by applying them on several m ulti-objective benchm ark instances of the traveling salesman problem. The results of the analysis have show n th at m ost of the considered MOACO algorithm s obtained better perform ances for m ore than tw o objectives and their perform ance depends slightly on the num ber of objectives, num ber of iterations and num ber of ants used.