新冠肺炎疫情演化情境下应急物资需求预测研究

Research on the medical allocation decisions in the context of COVID-19 epidemic evolution

根据新冠肺炎疫情演化过程中代表性事件、患病感染人数和疫情信息状态等特征,将其分为产生—爆发—高峰—衰退四个阶段.考虑湖北和除湖北以外的全国其他省市,绘制四个阶段下这两个区域每日新增患病感染人数图,总结每个阶段的特征.在假设患病感染者传染概率保持不变的前提下,基于传染病模型建立医疗物资时变需求模型;在患病感染者传染概率不断更新的基础上,采用贝叶斯序贯决策模型制定物资配置决策.比较两种模型下物资配置决策的差异,分析不同模型对计算结果的影响,在预测每日患病感染人数和实际每日患病感染人数情况下,贝叶斯模型的与传统需求模型的误差均小于时变需求模型.

According to the characteristics of information,number of infected persons and representative events during the evolution of COVID-19 epidemic,it is divided into four stages:emergence,outbreak,peak,and decline.Taking Hubei Province and other provinces of China except Hubei Province into consideration,the daily number of new infected person in these two parts under four stages was plotted.The characteristics of the four stage were summarized.Supposing that the transmission probability of infected persons remained unchanged,a time-varying demand model for medical supplies was established based on the SEIR epidemic model.Besides,considering that the transmission probability of infected persons is updated continuously,the Bayesian sequential decision model was adopted to make material allocation decision.The differences of material allocation decision between the two models were compared,and the influences of different models on the results were anlyzed.The errors of the Bayesian model and the traditional model are all smaller than those of the time-varing model for both the predicted and actual daily number of infections.