运力受限的应急物资动态调度模型及算法

针对大规模突发事件下应急物资动态调度问题, 把灾民的损失用物资未满足度的形式进行量化, 构建以最小化灾民损失和车辆调度费用为目标的混合整数规划模型, 以对运力受限情况下的救 援车辆路径选择和应急物资分配进行综合决策. 利用分层求解策略缩小问题的求解空间, 改进 基于客户点的编码方式, 设计了针对此问题的遗传算法. 代入实验数据得出了物资集散地不同 运力下的应急物资动态调度方案, 进而确定了最优运力配置量, 与现实情况下的配送方案比较 表明此模型和算法是有效的.

Model and algorithm of relief materials dynamic scheduling without sufficient vehicle quantity

This paper focused on relief materials dynamic scheduling after large emergency. We quantified the loss of victims in terms of unmet needs, and formulated a mixed integer programming model in order to minimize victims' loss and vehicle scheduling expense. The main decisions were determining the delivery routes and allocating the relief supplies from relief distribution center (RDC) to save points without sufficient vehicle quantity. We used the hierarchical thinking to reduce solution space, improved the coding method and designed the genetic algorithm for this problem. Finally, experimental results show the dynamic scheduling scheme under different vehicle quantity, and determine the optimal vehicle quantity in RDC through comparative analysis, the validity of this model and algorithm is verified combining with realistic distribution scheme.