Hybrid Ant Colony Algorithm Based on Scale Compression

To improve performance of ant colony algorithm when solving large-scale TSP problem, a hybrid ant colony algorithm based on scale compression is proposed. First we use genetic algorithm to generate a suboptimal solution set and calculate their intersection. By eliminating all cities mapped by the elements among the intersection in the primal TSP problem, we convert the original problem into a new one with smaller scale. In addition, we design a new optimal state transition rule based on regional characteristic of optimal solutions to accelerate convergence speed. Simulation results show our approach possess high searching ability and excellent convergence performance.