多车场动态路径问题的自适应量子蚁群算法

Self-adaptive quantum ant colony algorithm for solving multi depot dynamic vehicle scheduling problem

针对物流配送过程中存在的多配送中心动态需求车辆调度问题即多车场动态车辆调度问题(MDDVRP),提出了一种自适应量子蚁群算法(SAQACA),用于最小化路径。根据量子的相位编码方式,提出了对蚁群的信息素矩阵进行直接编码,进而实现由量子旋转门更新完成蚂蚁移动;根据搜索点的量子相位特点及目标函数的变化率,提出了一种自适应量子旋转门更新方式,进而提高了算法的全局搜索深度;引入基于两元素搜索策略的局部搜索方法提高了算法的局部优化能力,从而对可行解进行改进。仿真实验与算法比较验证了所提算法的有效性和优越性。

Aiming at multi depot dynamic vehicle routing problem（MDDVRP） existed in the process of logistics distribution,self-adaptive quantum ant colony algorithm（SAQACA） is proposed to minimize the distribution cost. According to the quantum phase encoding method,directly encoding of pheromone matrix of ant colony is presented to complete the movement of ants. According to quantum phase characteristics of search point and object functions change rate,a mode of adaptive quantum rotation gate is presented,to enhance global search depth,the local search method based on the principle of the two elements is introduced to enhance local optimization ability,so as to improve feasible solution. Simulation experiments and algorithm comparison demonstrate the effectiveness and the superiority of proposed algorithm.