不确定情景下考虑灾民感知的应急物资调度研究  

A study on the dispatch of emergency supplies under uncertainty scenarios considering the perception of disaster victims

作　　者：王付宇[1,2] 葛雪飞 王欣蕊 葛琬琪 李艳[1] WANG Fuyu;GE Xuefei;WANG Xinrui;GE Wanqi;LI Yan(School of Management and Engineering,Anhui University of Technology,Maanshan 243032,Anhui,China;Key Laboratory of Multidisciplinary Management and Control of Complex Systems of Anhui Higher Education Institutes(Anhui University of Technology),Maanshan 243002,Anhui,China;School of Pharmacy,Anhui Xinhua University,Hefei 230088,China)

机构地区：[1]安徽工业大学管理科学与工程学院,安徽马鞍山243032 [2]复杂系统多学科管理与控制安徽普通高校重点实验室(安徽工业大学),安徽马鞍山243002 [3]安徽新华学院药学院,合肥230088

出　　处：《安全与环境学报》2024年第5期1965-1976,共12页Journal of Safety and Environment

基　　金：国家自然科学基金项目(71872002,72274001);安徽省高校人文社科研究重大项目(SK2020ZD16)。

摘　　要：为解决大规模突发灾害给人民带来的生理与心理痛楚问题,考虑模糊需求情景下灾区道路受损、物资相对短缺、灾区需求紧迫度差异等因素,同时考虑灾民有限理性下物资竞争心理,运用前景理论刻画灾民对物资分配、运抵时间的综合感知,以灾区运输时间感知满意度最大、物资分配感知损失最小、运输成本最小为目标构建应急物资调度多目标优化模型,设计改进灰狼优化算法(Grey Wolf Optimizer,GWO)求解,引入混沌反向学习、差分进化、非线性收敛等策略实现对GWO算法的改进,并以2008年四川地震案例数据展开分析验证,依据模糊逻辑加权法选择合适的应急调度方案。研究表明,该模型可合理衡量有限理性下灾民综合感知,改进算法能够得出更加公平高效的调度方案,有效解决了灾后模糊需求情景下应急物资调度问题。Large-scale sudden disasters bring enormous physiological and psychological pain to people.This article first considers factors such as road damage,the relative shortage of materials,and differences in demand urgency in disaster areas under fuzzy demand scenarios.Besides,to describe the material competition psychology of the disaster victims under bounded rationality,the prospect theory is used to describe the comprehensive perception of the disaster victims on the material distribution and arrival time.The multi-objective optimization model of emergency material scheduling is constructed to maximize satisfaction of the transportation time perception in the disaster area,minimize the loss of the material distribution perception,and minimize the transportation cost.Then we design and improve the Grey Wolf Optimizer(GWO)to solve the model,introducing strategies such as chaotic reverse learning,differential evolution,and nonlinear convergence to improve the GWO algorithm.A two-dimensional matrix real number encoding method is selected to represent the material supply situation between supply and demand points and use a chaotic reverse learning strategy to sequentially execute chaotic mapping and reverse learning rules on the initial population.Nonlinear transformations on parameters are used to select optimal solutions from the constructed population,improve the quality of the initial population,and accelerate the convergence process.Based on the multi-objective optimization model established in this article,non-inferior solutions are obtained,and a fuzzy logic weighting method is designed to assist in selecting the scheduling plan in the event of sudden disasters.Finally,we conduct an analysis and validation based on the 2008 Sichuan earthquake case data and select the appropriate emergency dispatch plan using the fuzzy logic weighting method.The results show that the model can reasonably measure the comprehensive perception of victims under Bounded rationality,and the improved algorithm can get a more fair and eff

关 键 词：公共安全 感知满意度 模糊需求 应急物资分配 改进灰狼优化算法(GWO) 

分 类 号：X956[环境科学与工程—安全科学]