Optimization Approach Based on Immigration Strategies for Symmetric Traveling Salesman Problem

Abstract

The Traveling Salesman Problem (TSP) is a combinatorial optimization problem of great importance which continues to interest several researchers in order to develop methods to achieve an optimal solution. Genetic algorithms (GAs) as meta-heuristic methods have been widely applied to this problem. Inspired by biological phenomena, we introduce two immigration operators, random immigration and structured memory immigration, forming two different algorithms. The performance of these algorithms is evaluated using benchmark datasets of symmetric TSP from TSPLIB library. The results of the proposed algorithms are compared with the standard genetic algorithm showing that the proposed algorithms improve the performance of GA in solving TSP problem effectively and specifically with the developed structured memory immigration.