An Inspired Algorithm for Solving Competitive Travelling Salesmen Problem

Abstract

Competitive travelling salesman problem (CTSP) is a combinatorial optimization problem in which number of salesmen compete among themselves to fined optimal solution with larger benefit and shortest path. Despite the importance of this problem and its many applications in real life, a few algorithms have been proposed to address this problem. Consequently, the need to either improve the existing algorithms or utilize a new algorithm is still necessary. In the last decades, the nature inspired algorithms, which are seek inspiration from nature and biology phenomena, have been the goal of numerous studies in the most scientific fields, especially in operating research and artificial intelligence. The swarm intelligence describe as an active research area in the developments of new algorithms inspired by nature. One of the recent swarm intelligence algorithms is salp swarm algorithm (SSA). This algorithm is characterized as being simple and flexible, so it motivates scholars to conduct several modifications to improve its performance. But, as any population based metaheuristics, SSA suffers from the slow convergence due to its weak ability to exploit the search space. Thus, this paper proposes enhancing SSA to handle CTSP by utilizing its strong exploring ability and enhancing its exploitation ability. This enhancement achieves via hybridizing the SSA with a single-based meta-heuristics (SBHs) which have strong exploitation ability. In this hybridization, the SSA will be responsible for exploration and the SBH will be responsible for exploitation. The adopted algorithms are applied on CTSP benchmark to test their validity. Results demonstrated that preserving the balance between exploration and exploitation during the search have significant impact on the SSA efficiency. Thus, we concluded that the proposed hybridization managed to improve the effectiveness of SSA in getting good quality solutions.