Using PSO to Improve Ant Colony Optimization Algorithm

Ant colony optimization (ACO) is a swarm intelligence algorithm and it has been successfully applied to several NP-hard combinatorial problems such as traveling salesman, quadratic assignment problem (QAP), job-shop scheduling, vehicle routing and telecommunication networks. Howere, the ants' solutions are not guaranteed to be optimal with respect to local changes. In this paper, an improved ACO algorithm is proposed. Particle swarm optimization (PSO) has been applied to improve the performances of ACO. ACO is firstly used to find optimal solutions. Then PSO is used to optimize local optimal solutions searched by ACO. In order to check the performance of the proposed method, the proposed algorithm is utilized to solve QAP. The improved ACO algorithm and ACO algorithm are respectively implemented on some instances extracted from QAPLIB. The experimental results demonstrate that the improved ACO algorithm has better performance in terms of the quality of the returned solution than the original ones.