应对自然灾害的应急资源布局模型与算法

Optimal resource location and allocation model for natural disasters

以地震为背景,考虑灾害发生时需求不确定的条件,建立基于情景分析的随机整数规划模型,解决针对自然灾害的应急资源布局问题.将灾害发生后的情景划分成两个阶段的随机事件,前一个随机事件表示灾害刚发生后震源位置、震级大小的信息;后一个表示当震源和震级的信息确定后,各个灾区的需求量.利用有限个情景表示不确定性数据.通过松弛非预期约束,将松弛问题按照情景分解,并利用分支定界算法求解Lagrange松弛问题.数值试验表明,建立的模型是实际可行的,而且算法也是有效的.

We propose a stochastic programming model to determine a rescue resource layout system for earthquake disasters. Randomness is represented by a finite sample of scenarios with two stage stochastic evens. One event is information about epicenter and magnitude and the other is resource requirements of demand points. The proposed model is solved using a Dual Decomposition algorithm, which makes use of special structure of the model. Finally, we propose a case study to highlight efficiency of the proposed solution strategy.