The modification of hybrid method of ant colony optimization, particle swarm optimization and 3-OPT algorithm in traveling salesman problem

Abstract

The traveling salesman problem (TSP) is a famous problem in finding the shortest tour to visit every vertex exactly once, except the first vertex, given a set of vertices. This paper discusses three modification methods to solve TSP by combining Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO) and 3-Opt Algorithm. The ACO is used to find the solution of TSP, in which the PSO is implemented to find the best value of parameters α and β that are used in ACO.In order to reduce the total of tour length from the feasible solution obtained by ACO, then the 3-Opt will be used. In the first modification, the 3-Opt is used to reduce the total tour length from the feasible solutions obtained at each iteration, meanwhile, as the second modification, 3-Opt is used to reduce the total tour length from the entire solution obtained at every iteration. In the third modification, 3-Opt is used to reduce the total tour length from different solutions obtained at each iteration. Results are tested using 6 benchmark problems taken from TSPLIB by calculating the relative error to the best known solution as well as the running time. Among those modifications, only the second and third modification give satisfactory results except the second one needs more execution time compare to the third modifications.