不确定与损毁情景下可靠性设施选址鲁棒优化模型与算法研究

设施选址是长期战略性决策问题,选址决策面临各种不确定因素,设计一个可靠的选址网络具有重要的战略意义.本文同时考虑需求的不确定性及设施可能损毁的情景,扩展无容量限制的固定费用可靠性选址模型,建立不确定与损毁情景下服务能力有限的可靠性设施选址鲁棒优化模型.基于Bertsimas和Sim鲁棒优化方法提出一个新的混合整数规划模型,通过引入辅助变量和对偶变换实现非线性鲁棒优化模型转化为鲁棒对应模型,提出蝙蝠算法(BA)对模型予以求解,并通过算例仿真验证模型和算法的可行性,为设施选址决策提供模型和方法设计.

Robust optimization model and algorithm for reliability facility location under uncertainty and failure scenarios

Facility location is a long-term strategic decision problem, but the location decision-making is confronted with various uncertain factors. Therefore, it is of strategic importance to design a reliable location network. This paper considers both demand uncertainty and facility failures, extends uncapacitated fix-charge location problem to establish reliability facility location robust optimization model under uncertainty and failure scenarios. A new mixed integer programming model is proposed based on Bertsimas and Sim robust method, the nonlinear robust optimization model is transformed into a robust correspondence model by introducing the auxiliary variable and dual transformation. We develop bat algorithm (BA) to solve the model, our computational results show the feasibility of the presented model and algorithm. This paper provides model and method design for facility location decision.