车船协同的洪灾被困人员搜救路径鲁棒优化  

Robust optimization of vehicle-vessel coordinated search and rescue routing for flood-stranded individuals

作　　者：周愉峰[1] 龚英[1] 刘晓聪 何珺阳 ZHOU Yufeng;GONG Ying;LIU Xiaocong;HE Junyang(School of Management Science and Engineering,Chongqing Technology and Business University,Chongqing 400067,China)

机构地区：[1]重庆工商大学管理科学与工程学院,重庆400067

出　　处：《安全与环境学报》2025年第4期1455-1465,共11页Journal of Safety and Environment

基　　金：国家社会科学基金项目(23XGL039)。

摘　　要：优化车船协同搜救路径,以提高洪灾被困人员搜救效率。首先,引入洪水模拟系统和匮乏成本函数,以总匮乏成本最小为目标,采用0-1混合整数规划,构建救援时间不确定且有公平性约束的被困人员搜救路径优化模型。之后,引入鲁棒优化方法,将模型转化为等价的鲁棒优化模型。再根据模型约束特征设计若干修复算子,提出一种改进的禁忌搜索(Improved Tabu Search, ITS)算法。最后,设计两组算例验证模型和算法的有效性和可靠性。结果表明:考虑被困人员救援时间的不确定性有利于决策者优化救援方案;ITS性能优于传统禁忌搜索算法、遗传算法与模拟退火算法,对真实算例的平均优化效果分别为3.10%、15.76%与10.27%。研究成果可为应急管理部门优化被困人员搜救策略提供决策参考。This paper explores the optimization of coordinated search and rescue routes for vehicles and vessels to improve the efficiency of rescuing individuals trapped by floods.First,a flood simulation system and a deprivation cost function are introduced to construct an optimization model for search and rescue routes.The model accounts for uncertainties in rescue times and incorporates fairness constraints to ensure equitable resource allocation.The objective of this model is to minimize the total deprivation cost.It is formulated as a 0-1 mixed-integer programming problem.To account for uncertainties,robust control parameters are introduced,transforming the model into an equivalent robust optimization problem.Based on the model s constraint characteristics,several tailored repair operators were developed to enhance the search performance of the algorithm.An Improved Tabu Search(ITS)algorithm was proposed,incorporating these repair operators.To validate the model and the algorithm,two sets of instances were created,and a sensitivity analysis was conducted on key parameters.The first set of instances was based on the catastrophic flooding caused by the“7.20”heavy rainfall disaster in Zhengzhou,Henan Province,in 2021.The second set comprised five simulated scenarios of varying scales.Numerical analysis results confirm the effectiveness and reliability of both the model and the algorithm.Key findings are as follows:(1)Incorporating uncertainties in the rescue time of trapped individuals enables decision-makers to optimize rescue plans more effectively,allowing for decisions that are better aligned with specific preferences.(2)The ITS algorithm outperforms traditional methods,including the standard Tabu Search,Genetic Algorithm,and Simulated Annealing,achieving optimization improvements of 3.10%,15.76%,and 10.27%,respectively,in the real case study.(3)A critical threshold exists in the escalation of disaster severity;within this range,disaster conditions deteriorate rapidly as rainfall increases.Governments should leve

关 键 词：公共安全 洪灾救援 车船协同 车辆路径问题 禁忌搜索算法 

分 类 号：X43[环境科学与工程—灾害防治] X959