An Efficient Combined Meta-Heuristic Algorithm for Solving the Traveling Salesman Problem

Abstract

The traveling salesman problem (TSP) is one of the most important NP-hard Problems and probably the most famous and extensively studied problem in the field of combinatorial optimization. In this problem, a salesman is required to visit each of n given nodes once and only once, starting from any node and returning to the original place of departure. This paper presents an efficient evolutionary optimization algorithm developed through combining imperialist competitive algorithm and lin-kernighan algorithm called (MICALK) in order to solve the TSP. The MICALK is tested on 44 TSP instances involving from 24 to 1655 nodes from the literature so that 26 best known solutions of the benchmark problem are also found by our algorithm. Furthermore, the performance of MICALK is compared with several metaheuristic algorithms, including GA, BA, IBA, ICA, GSAP, ABO, PSO and BCO on 32 instances from TSPLIB. The results indicate that the MICALK performs well and is quite competitive with the above algorithms.