【摘要】 针对含有连续分布随机参数的回收物流网络优化设计问题,结合抽样理论,建立了由样本数量决定求解效率的二阶段随机规划模型.指出适量小样本对应的模型最优值必然是实际最优值的下界,提出了基于大样本分析的物流网络稳健性评价方法以及实际最优值上界的确定方法.给出模型求解的混合遗传算法,并总结了物流网络的优化设计步骤.通过具体算例说明了模型及其算法在设计决策中的应用.更多还原

【Abstract】 With the help of sampling theroy,a two-stage stochastic programming model is developed to solve the returned logistics network optimal design problems with continuous distribution stochastic parameters,whose computation efficiency is determined by sample quantity.And the model’s optimal values resulting from appropriately small scale samples are always lower bounds of the true optimal value.Network robust estimation approach and true optimal value’s upper bound judgement method are presented based on corresponding analysis to large scale samples.Mixed genetic algorithm of the model is given,and the network optimal design steps are summarized.An example shows the application of the proposed model and its algorithm in design decision-making.更多还原