A Swarm-Intelligence Based Algorithm for Solving Traveling Salesman Problem

Ai, The Jin (2011) A Swarm-Intelligence Based Algorithm for Solving Traveling Salesman Problem. Australian Society For Operations Research (ASOR), 30 (4). pp. 2-10. ISSN 1446-6678

Preview	Text Paper 10 ASOR SI for TSP (short).pdf Download (4MB) | Preview
Preview	Text Paper 10 Peer Riview.pdf Download (554kB) | Preview
Preview	Text Turnitin 10.pdf Download (3MB) | Preview

Abstract

The traveling salesman problem (TSP) plays an important role in the field of physical distribution and logistics. This problem is generally defined as the problem for dete11nining the sequence of the cities to be visited by a traveling salesman such that the operational cost of the traveling is minimized. I n recent two decades, the swarm-intelligence ph ilosophy has been emerging into some optimization algorithms such as ant colony optimization (ACO) and iL variants, also particle swarm optimization (PSO) and its variants. Both ACO and PSO, despite their strengths and weaknesses, are reported successfully being applied to solve the TSP. This paper tries 10 develop a new solution methodology for solving the travel ing salesman problem using different point of view on the swarm-intelligence philosophy in the form of a new proposed algorithm.The expectations of the new proposed algorithm are reducing the weakness and increasing the strength of its application. The performanc.e of the proposed algorithm is then evaluated usi11g severa l benchmark datasets for the traveling salesman problem. The computacional results show chat che proposed algorithm using specific settings is able to find good solution of the corresponding traveling salesman problem instance. It is also sti ll possible to improve the result by fine tuning the algorithm and add ing an effe.ctive local search method.

Actions (login required)

View Item