关于配送线路的混合禁忌搜索自整定仿真研究

Research on Hybrid Tabu Search Self-Tuning Simulation of Distribution Line

物流运输网络中的线路和节点状态具有动态变化的特性,要求配送车辆必须根据网络变化对运输线路采取实时调整。为了更加准确快速的完成最优配送线路规划,提出了混合禁忌搜索自整定方法。方法首先将运输成本作为配送线路的寻优目标,根据距离与时间等因素,设计了关于配送线路的软时间窗口模型与约束条件。然后利用蚁群作为寻优的基础算法,对每一条配送线路标记信息素。考虑到蚁群算法的局部解缺陷,引入混合禁忌搜索,在迭代处理时加入信息素因子,用于扰动信息素寻优的结果。同时对每次迭代出的最优解设计了优化机制,用于更新信息素和约束。最后通过仿真,证明了提出的混合禁忌搜索方法具有良好的寻优性能,优化得到的配送线路符合距离、时间、成本的综合需求,提高配送效率的同时,有效抑制了运输成本的增长,能够友好的应对物流运输网的动态变化与客户数据的急剧增加。

The line and node states in the logistics transportation network have the characteristics of dynamic changes,which requires the distribution vehicles to adjust the transportation line in real time according to the network changes.In order to complete the optimal distribution route planning more accurately and quickly,a hybrid tabu search self-tuning method is proposed.Firstly,the transportation cost was regarded as the optimization target of the distribution route.According to the factors of distance and time,the soft time window model and constraints of the dis⁃tribution route were designed.Then ant colony was used as the basic algorithm to mark pheromones for each distribu⁃tion line.Considering the local solution defect of ant colony algorithm,the hybrid tabu search was introduced.The pheromone factor was added to the iterative processing to disturb the result of pheromone optimization.At the same time,the optimization mechanism was designed to update pheromones and constraints.Finally,the simulation results show that the proposed hybrid tabu search method has good performance,and the optimized distribution lines can meet the comprehensive requirements of distance,time and cost.At the same time of improving the efficiency of dis⁃tribution,it can effectively restrain the growth of transportation cost and deal with the dynamic change of logistics transportation network and the sharp increase of customer data.