Self-Adaptive Social Spider Algorithm optimization for Solving Travelling Salesman Problem

Abstract

In this paper, a new method is presented based on Social Spider Algorithm (SSA) for solving travelling salesman problem (TSP). Since the SSA is applied to continuous problem and TSP is discrete NP-hard problem, adaptive version of the SSA is introduced that called Adaptive SSA (DSSA). The DSSA algorithm was implemented on 36 instances of TSPLIB benchmarks (a library of sample instances for the TSP) that include symmetric and asymmetric traveling salesman problems. In order to implement the DSSA, MATLAB 2017 was used. After simulation, DSSA found the optimal solution for 24 instances out of 36 datasets. The simulation results showed that DSSA is superior to other algorithms for solving both the TSP and ATSP problems. We propose the Adaptive Social Spider Algorithm (DSSA) based on SSA for solving adaptive optimization problems such as TSP problem. The DSSA is main contribution of this paper. In this section, the changes of original SSA are described to create adaptive version of SSA.