基于多层次蚁群算法的快递路径优化

Optimizing route of logistics based on multi-level ant colony algorithm

文中根据快递路径的特殊性,针对城内交通和城市间交通的差异性,设计出多层次蚁群算法。在传统蚁群算法的基础上,将网络节点划分为不同层次,并在不同层次内采用不同的状态转移策略,选用不同的启发函数;同时,为了提高算法的收敛速度,并且使其不容易陷入局部最优,该算法对信息素的更新策略进行了调整。利用多层次蚁群算法实现了快递网络中成本和时间折中考虑的路径选择问题,通过与基本蚁群算法找出的时间最优和成本最优路径进行比较,证明了该算法的可行性和有效性。

Due to the particularity of delivery routes and the differences between urban transport and intercity transport,a multi-level ant colony algorithm was proposed. O n the basis of the traditional ant colony algorithm,the nodes was divided into different levels. Then different state transition strategies and heuristic functions were applied in different levels. In order to improve the convergence speed and avoid the local optimum,the rule of updating the pheromone was improved. The improved algorithm was then applied on logistics delivery problems considering both logistics time and cost. The time-least paths and cost-least paths were found by the traditional ant colony algorithm. Then the feasibility and effectiveness of the improved algorithm were proved by comparing the results obtained by multi-level ant colony algorithm and traditional ant colony algorithm.