A dual population parallel ant colony optimization algorithm for solving the traveling salesman problem

In allusion to the phenomenon of stagnation and precocity during evolution in ant colony optimization (ACO) algorithm, this paper proposed a dual population parallel ant colony optimization (DPPACO) algorithm, which was applied to the traveling salesman problem. The DPPACO algorithm separated the ants into soldier ant population and worker ant population which evolve separately by parallel method and exchanges information timely. The dynamic equilibrium between solution diversity and convergence speed is achieved by using the effect of the soldier ant’s distribution to worker ants’ movement choice. The DPPACO algorithm can enlarge searching range and avoid local minimum, prevent local convergence caused by misbalance of the pheromone and can improve the searching performance of the algorithm effectively. The proposed algorithm is applied in the traveling salesman problem by using the 17 data sets obtained from the TSPLIB. We compare the experimental results of the proposed DPPACO method with the traditional methods. The experimental results demonstrate that the proposed algorithm has a better global searching ability, higher convergence speed and solution diversity.