考虑优先级和时间窗约束的应急物资调配模型  

Emergency Material Deployment Model Considering Priority and Time Window Constraints

作　　者：王慧丽 乔咏艺 WANG Hui-li;QIAO Yong-yi(School of Statistics,Xi an University of Finance and Economics,Xi an 710100,China)

机构地区：[1]西安财经大学统计学院,西安710100

出　　处：《科学技术与工程》2024年第12期5069-5075,共7页Science Technology and Engineering

基　　金：陕西省创新能力支撑计划(2022KJXX-36);陕西省教育厅专项科学计划项目(20JG010);陕西高校青年创新团队(2023128)。

摘　　要：突发公共卫生事件的不确定性和突发性特点,使得高效精准的应急物资调配方案显得尤为重要。构建突发公共卫生事件背景下生活物资配送中转站的选址模型,并考虑特殊事件的发生,改进优先级系数和时间窗的计算,以运输距离、违反车辆容量与时间窗约束的惩罚成本最小为多目标,构建物资配送路径规划模型,通过K-means聚类算法解决选址问题,设置遗传算法和大邻域搜索算法的混合遗传算法求解最优配送路径。最后以长春市朝阳区的物资配送为例进行实证分析。结果表明:可以将200个小区聚类为60个物资需求站点进行配送,需要26辆车将物资从调配中心转运到需求站点,并得到4类最优的物资配送方案,为解决突发公共卫生事件下应急物资配置提供了新的思路。The uncertainty and sudden nature of public health emergencies make efficient and accurate emergency material allocation plans particularly important.A site selection model of daily necessities distribution transfer station under the background of public health emergencies was constructed.Considering the occurrence of special events and updating priority coefficients and time window calculations,a multi-objective delivery path planning model was established using transportation distance,penalty cost for violating vehicle capacity and time window constraints as the optimization goals.The site selection problem was solved by the K-means clustering algorithm and the optimal delivery path was obtained using a hybrid genetic algorithm combining genetic algorithm and large neighborhood search algorithm.Finally,the example of material distribution in the Chaoyang District of Changchun was used to empirical analysis.The results indicate that 200 neighborhoods can be clustered into 60 material demand sites for distribution.It requires 26 vehicles to transport the materials from the distribution center to the demand sites and four types of optimal material distribution schemes are obtained,which provides a new idea for solving the emergency material allocation under public health emergencies.

关 键 词：物资调配 路径规划 多目标 K-MEANS聚类算法 混合遗传算法 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]