Abstract: The limitation of existing transportation route optimization was that its decision mainly relied on experience． Uncertain factors such as network reliability，capacity liability，time，transport demand，transport loss in rescue transportation were analyzed． Among them，three priority objectives，namely time，reliability and loss were determined． Multi-objective stochastic programming model with network diagram theory for rescue transportation was built． Genetic algorithm was applied to calculating． Each parameter could be illustrated by concrete example and validity of the model was verified through actual simulation．

Key words: natural disaster, rescue transportation, path optimization, genetic algorithm

摘要： 针对目前运输路径优化模型主要依靠经验法做出决策判断的缺陷，分析了灾后救援运输过程中网络可靠度、 容量可靠度、运输时间、运输物资需求量、运输损耗等各种不确定性因素;分别以时间目标、可靠性目标、损耗目标 为3个优先级，运用网络图理论构建了灾后应急救援运输路径优化的多目标随机规划模型，并运用遗传算法探讨 模型的求解方法;通过具体的案例来说明模型中各个参数，并通过实际的模拟计算来验证该模型的合理性和有效 性。

关键词: 自然灾害, 救援运输, 路径优化, 遗传算法