具有同时集送货需求车辆路径问题的混沌量子进化算法研究

Chaos quantum evolutionary algorithm for vehicle routing problem with simultaneous delivery and pickup

针对量子进化算法中旋转角取值的离散性使其解空间的搜索具有跳跃性,提出了基于混沌理论的精英均值计算旋转角算法,并将其应用于具有同时集送货需求车辆路径问题的求解.在理论上分析了解的强可行和弱可行条件的基础上,使用启发式算子对解进行改进.通过仿真实验与其他算法进行了比较,仿真结果表明所提出算法是求解此类问题的有效方法.

For the discrete value of the rotation gate,the quantum-inspired evolutionary algorithm （QEA） has jump phenomena in the search space. To improve the QEA,a hybrid algorithm with computing the rotation gate using elite mean values based on chaos theory is presented,which is applied to vehicle routing problem with simultaneous delivery and pickup （VRPSDP）. The solution of VRPSDP is investigated,and the qualification of strong feasible solution and puny feasible solution is analyzed theoretically. An efficient population initialization based on nearest insertion algorithm （NIA） and chaos function is proposed to generate an initial population with certain quality and diversity. Simulation results and comparisons show the effectiveness of the proposed algorithm.