Discrete Salp Swarm Algorithm for Euclidean Travelling Salesman Problem

Karuna Panwar, Kusum Deep

Abstract: The Salp Swarm Algorithm (SSA) is one of the recently proposed swarm intelligence-based algorithms, which finds its motivation in the swarming behaviour of salps when navigating and foraging in the ocean. The SSA is essentially derived to solve optimization problems that have continuous search space. SSA is a simple to implement and competitive algorithm that has been proven helpful in a variety of real-world applications. It has been explored over several optimization problems so far. In this study, an enhanced discrete version of SSA is proposed to solve the Travelling Salesman Problem (TSP). As TSP is a combinatorial optimization problem, the classical SSA is modified using swap, shift and symmetry operators for global exploration and local exploitation. Also, the 2-opt method has been incorporated with the proposed algorithm to improve the local search ability of the algorithm while addressing discrete problems. The proposed Discrete Salp Swarm Algorithm (DSSA) has evaluated over 45 TSP instances, and the computational results showed that it is a promising algorithm. To assess its performance, the proposed algorithm’s results are compared with well-known algorithms such as Genetic Algorithm, Artificial Bee Colony, Spider Monkey Algorithm, Jaya Algorithm, Black Hole, Symbiotic Organism Search etc. The proposed algorithm significantly outperformed these algorithms for a majority of TSP instances.