关键词: 多级设施中断, 需求不确定, 选址—库存模型, 供应链网络优化, 混合遗传算法

Abstract: Focusing on tree-shaped supply chain system consisting of plants,distribution centers and retailers,an appropriate network structrue design approach was developed.By considering demand uncertainty of the retailers and interruption risks of the plants and distribution centers,a supply chain location-inventory decision model with the goal of minimizing expected total cost of the system was established.The model could be used to determine the locations,number and storage capacities of the distribution centers,as well as the replenishment/delivery assignment level of each plant/distribution center to each distribution center/retailer.To rapidly solve the proposed model,a hybrid genetic algorithm with a specific cost analysis method embedded in the genetic algorithm framework was designed,and the operation ideas and implementation steps of the algorithm were given.In addition,the experiments were constructed to verify the effectiveness of the model and its algorithm,and the deduction was obtained simultaneously that under uncertain demand environment,the decision result by considering multi-echelon facility interruptions was better than those without considering facility interruptions or considering only single-echelon facility interruptions.

Key words: multi-echelon facility disruption, uncertain demand, location-inventory model, supply chain network optimization, hybrid genetic algorithm