H-ACO Algorithm for the VRPTW with Re-used Vehicles

In this paper, the VRPTWRV (VRPTW with Re-used Vehicle) was presented and a mathematical programming model with multiple objects was constructed correspondingly. The heuristic information and Pseudo-random probability selection rule are improved according to the principle of serving customers with earlier service beginning time, shorter service time and earlier ending time firstly based on Ant Colony System. Pheromone updating strategy is designed by combining the MAX-MIN and Rank-based Ant System. By adding a local search strategy into the improved Ant colony algorithm, a H-ACO algorithm is presented for the VRPTWRV. At last the algorithm is tested on the 8 data groups of Solmon C2. Numerical results show that the H-ACO can obtain the satisfying solutions quickly, and the global convergence performance of the algorithm is good.