不确定需求条件下大规模抗震救灾应急动员优化方法  

Mobilization optimization method for large-scale emergency earthquake disaster relief with uncertain demands

作　　者：刘亚杰[1] 王文峰[2] 雷洪涛[1] 郭波[1] 

机构地区：[1]国防科技大学信息系统与管理学院,长沙410073 [2]空军装备研究院,北京100085

出　　处：《系统工程理论与实践》2013年第11期2910-2919,共10页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(71371181;91024006);中国博士后科学基金(2012M521918)

摘　　要：地震灾害的突发性及预测困难性使抗震救灾动员工作的重要性得到凸显.针对我国主要地震带的分布特点,考虑震后初期动员阶段对救援品需求的不确定性和应急救援环境的特殊性,建立一个选址-分配随机混合整数规划模型,以确定抗震救灾动员阶段的配送设施选址、配送集散水平以及总体运输分配计划等决策问题,并基于模型的结构特点提出了一种基于Lagrangian松弛的快速求解算法.最后通过汶川地震救援的模拟数据实例对所建模型和所提算法的有效性进行了验证.Earthquakes are a special type of disaster that has been proved difficult to predict exactly while occur with abruptness, and these two attributes make the post-disaster mobilization work become a vital procedure. According to the main distribution of earthquake faults in our country, a facility location- allocation stochastic mixed integer model, which considered the uncertain demands of disaster areas and the special environment in disaster relief, was proposed to determine the facility locations of distribution centers, the initial inventory levels at these established facilities and the overall transportation plans from facilities to disaster areas. Furthermore, a heuristic algorithm based on Lagrangian relaxation idea was also developed according to the special structure of the proposed model. The efficiency of the proposed model and algorithm was validated by a data case that constructed from the simulative situation of the great Wenchuan earthquake.

关 键 词：应急决策 动员优化 随机规划 模型 算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]