A Revised EM-Like Algorithm + K-OPT Method for Solving the Traveling Salesman Problem

In this study, the main objective is to solve the traveling salesman problems (TSP). TSP belongs to a part of the combinatorial optimization problems and NP-complete problems. Though this problem is easy to describe and understand, when the problems of the dimensions are bigger, it becomes difficult and hard to solve. If we consider all feasible solutions, it may cost a lot of time and not be effective. In this study, we utilize the new meta-heuristic algorithm, called the electromagnetism-like algorithm (EM) which was proposed by Birbil and Fang [2003], to solve the TSP problem. Although EM has been used to solve the TSP, the results sometimes can not jump out the local optimum. Therefore, a revised EM-like algorithm + k-opt method is introduced and expects to obtain a better solution than the original EM