失效情景下基于Stackelberg博弈的溢油应急物资储备库选址优化模型

Stackelberg game based optimization model of oil spill emergency material reserve location in failure scenarios

为优化溢油应急物资储备库的选址,将溢油事故假想为攻击者的攻击后果,以Stackelberg主从博弈理论为基础,考虑应急物资储备库的失效情景,将政府视为先行决策者,攻击者作为跟随决策者建立双层规划模型。设计分支定界算法、蒙特卡洛随机模拟算法和遗传算法相结合的混合算法进行求解。案例结果表明:提出的模型可以实现对区域内风险点的完全覆盖以及对重大风险点的多重覆盖;在失效情景下,相较于传统覆盖选址模型,提出的模型的选址结果使得平均溢油损失减少2.4%,平均总成本减少0.4%,提高了应急系统的服务水平,验证了模型及算法的有效性。

In order to optimize the location of oil spill emergency material reserve warehouses,this paper assumes that the oil spill is the consequence of the attacker s attack.Based on the Stackelberg master-slave game theory,considering the failure scenarios of the emergency material reserve warehouses,the government is regarded as the first decision maker,the attacker is regarded as the follower to make decision,and a bi-level programming model is established.A mixed algorithm combining the branch and bound algorithm,the Monte Carlo stochastic simulation algorithm and the genetic algorithm are designed to solve the model.The case results show that:the proposed model can achieve full coverage of risk points within the region as well as multiple coverage of significant risk points;in the failure scenarios,compared with the traditional coverage location model,the location results of the proposed model make the average oil spill loss reduce by 2.4%and the average total cost reduce by 0.4%,which improves the service level of the emergency system and verifies the effectiveness of the model and the algorithm.