可多次配给的多层级应急调度模型与算法

Multiple Rationing Multi-level Emergency Material Dispatching Model and Its Algorithm

针对多层级的非线性连续消耗应急物资调度问题,以受灾点物资缺失损失最小和物资调度总成本最小为目标,构建由陆运总站、空运总站、分配中心和受灾点组成的多层级、多次数配给、多运输方式、可转运的应急物资调度模型.鉴于多目标应急物资调度模型的特点,提出了一种基于双变异策略的多目标差分进化算法.该算法能够通过反馈解来实现对不同个体采用不同的变异策略,从而有效平衡算法的探索性与开发性.通过仿真对比实验,验证了模型的合理性和算法的有效性.算例分析表明,该算法能够明显增加解的数量并提高算法的收敛性能.

In view of the characteristic of the multi-level nonlinear continuous-consumption emergency material dispatching problem,to decrease shortages of affected areas and reduce the cost of material dispatching,we establish a multi-level,multiple-rationing,multi-transportation mode,and transshipment-permitted emergency material dispatching model that includes land and air terminals,distribution centers,and affected areas.We propose a multi-objective differential evolution algorithm based on the double-mutation strategy for the multiobjective emergency material dispatching model.The algorithm can use different variation strategies for different individuals by using feedback solutions to effectively balance the exploration and exploitation of the algorithm.By using a simulation and a contrast experiment,we prove that the model is reasonable and that the algorithm is effective.An analysis of examples indicates that the algorithm can increase the solutions and enhance the convergence performance of the algorithm.