应急救援物资多目标分配与调度问题建模与求解

Modeling and solving multi-objective allocation-scheduling of emergency relief supplies

应急救援物资多目标分配和调度集成优化是灾害应急决策中的一个难点问题.为此,构建一种面向多储备点、多发放点、多种应急救援物资的并发分配与调度多目标优化模型,提出一种基于二维NSGA-II与蚁群优化的混合智能搜索算法.在所提算法中,设计二维二进制编码、个体修正策略以解决多发放点之间潜在的应急救援物资冲突,并改进蚁群优化中的信息素全局更新策略以实现多个调度路径的同时优化.对比实验结果表明,所提出的混合算法可以在合理的时间内给出更高质量的应急救援物资分配和调度方案.

The multi-objective allocation-scheduling of emergency relief supplies is a very challenging topic in the field of disaster emergency decision-making. Therefore, a multi-objective optimization model of allocation-scheduling is developed,taking reserve points, dispatch points, and emergency relief supplies altogether. A hybrid algorithm on the basis of NSGAII and ant colony optimization（ACO） is proposed for solving the problem. Then NSGA-II with two-dimensional binary chromosome encoding and the corresponding strategy for individual repairs are designed to solve the potential conflicts over emergency relief supplies among multiple dispatch points. Moreover, an improved strategy is presented for updating the global pheromone in ACO to search for different scheduling routes at the same time. Finally, the experimental results show that the proposed hybrid algorithm can obtain higher quality schemes of emergency relief supplies allocation-scheduling in a reasonable time.