基于长短期记忆网络的新冠防疫资源调度方法

COVID-19 epidemic prevention resource scheduling method based on Long Short⁃Term Memory network

为解决新型冠状病毒肺炎防控过程中存在的资源利用率低的问题,提出一种基于长短期记忆网络的新冠防疫资源调度方法。模型通过应用程序接口采集新型冠状病毒肺炎防疫的检测阳性患者、防疫人力资源、物资数据;采用长短期记忆网络,实现了日、周、月多时间尺度的新冠疫情新型冠状病毒肺炎情况的预测;在此基础上,计及新型冠状病毒肺炎防疫的人力、物资资源约束条件,根据新型冠状病毒肺炎蔓延趋势预测情况对新型冠状病毒肺炎防疫资源进行调度。在某城市进行了仿真验证,其新冠疫情新型冠状病毒肺炎防疫资源利用率为87.8%,说明所提方法能有效提高新型冠状病毒肺炎防疫资源利用率。

In order to solve the problem of low resource utilization in the prevention and control of novel coronavirus pneumonia,a resource scheduling method for COVID 19 epidemic prevention based on Long Short⁃Term Memory network was proposed.The model collects the data of positive patients,epidemic prevention human resources and materials of COVID-19 novel coronavirus epidemic prevention through the application program interface;The long⁃term and Long Short⁃Term Memory network was used to realize the prediction of COVID-19 novel coronavirus epidemic situation on a daily,weekly and monthly timescale.On this basis,taking into account the constraints of human and material resources for epidemic prevention of novel coronavirus,the epidemic prevention resources of novel coronavirus are scheduled according to the prediction of the spread trend of novel coronavirus.A simulation was conducted in a city,and the utilization rate of resources for COVID 19 novel coronavirus epidemic prevention was 87.8%.The proposed method can effectively improve the utilization rate of resources for epidemic prevention of novel coronavirus pneumonia.