Multi-Agent in Ant Colony Algorithm Approach for Solving Traveling Salesman Problem

Traveling salesman problem (TSP) is a very hard and classical optimization problem in the field of operations research, and often-used benchmark for new optimization techniques. This paper will to bring up multi-agent approach for solving the TSP based on data mining algorithm, for the extraction of knowledge from a large set of TSP. The proposed approach supports the distributed solving to the TSP. It divides into three-tier, the first tier is ant colony optimization agent; the second-tier is genetic algorithm agent; and the third tier is fast local searching agent. In using an ant colony algorithm (ACA) for the TSP, an attribute-oriented induction methodology was used to explore the relationship between an operations´ sequence and its attributes and a set of rules has been developed. These rules can duplicate the ACA´s performance on identical problems. Ultimately, the experimental results have shown that the proposed hybrid approach has good performance with respect to the quality of solution and the speed of computation.