数据驱动的疫情应急物流网络动态调整优化

突发疫情具有扩散边界模糊、情景动态时变等典型非结构化特征,相应的应急物流网络需要根据实际情况不断进行调整,才可以取得更好的救援效果.基于此,本文从数据驱动的视角,构建了一类创新的应急物流网络动态调整优化决策框架模型.其中,应急响应时间被划分为多个连续的决策周期,每个决策周期中蕴含了疫情扩散分析、应急物流网络设计、数据收集处理和参数调整更新等循环递进的4个环节.在该决策框架下,整个疫情的应急响应过程转化为数据学习与资源优化配置交互演进的协同决策过程.算例分析表明,本文所设计的决策框架模型能够为疫情应急管理提供许多实时有效的政策调整建议,也可为其它突发事件的应急应对提供有益的决策参考.

Dynamic adjustment method for optimizing epidemic-logistics network based on data-driven

To improve the effect of emergency response in epidemic controlling, the corresponding logistics network should be adjusted dynamically because an unexpected epidemic outbreak has several typical unstructured features, including the fuzzy boundary and time-varying decision scenarios. In this paper, an innovative decision framework for optimizing the epidemic-logistics network based on data-driven is proposed. The whole emergency response time is divided to be multiple and continuous cycles. Emergency response process in each decision-making cycle involves four steps, which are epidemic dynamics analysis, emergency distribution network design, data collection, and parameters adjustment. Under this new decision framework, the entire emergency response process can be converted to an interactive evolution process of data learning and resource optimization. Numerical tests demonstrate that the proposed new decision framework can provide several real-time and effective policies for controlling an unexpected epidemic outbreak. Moreover, it also provides useful decision-making reference for other emergencies.