Research on Vehicle Routing Problem with Time Windows Based on Quantum-behaved Particle Swarm Optimization

摘要

In view of the fact that utilizing Particle Swarm Optimization (PSO) to solve Vehicle Routing Problem with Time Windows (VRPTW) is easy to converge to local optimal solution, we propose Quantum-Behaved Particle Swarm Optimization (QPSO) to solve VRPTW (QPSO-VRPTW). In the first step, Low Dimension Real-coded method (LDC) is used to code particle's state, which gives full play to advantages of QPSO. The second step is to re-optimize the route by nearest insertion method and 2-Opt method, which can improve quantity of routes. In the last step, we utilize QPSO to update the particle state and search for the optimal solution. The experiment results show that compared with the PSO-VRPTW, the QPSO-VRPTW has better performance in convergence velocity and global searching ability.