首页|重大传染病疫情下应急医疗物资需求预测和配置研究

重大传染病疫情下应急医疗物资需求预测和配置研究

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
为了科学合理地进行应急医疗物资配置，提高重大传染病疫情防控效率，根据疫情演化不同阶段的特点开展应急医疗物资需求预测和配置研究。首先，根据疫情数据特征，提出传染病模型 SEIR(Susceptible Exposed Infectious Recovered)和长短期记忆(Long Short-Term Memory，LSTM)网络相结合的模型(SEIR-LSTM)预测各需求点的应急医疗物资需求量，该方法利用LSTM对时间序列数据良好的学习能力预测感染率，输入SEIR模型提高预测准确率。然后，根据传染病疫情演化关键阶段的特点，考虑物资配送成本、需求紧迫度和分配公平性等因素构建分阶段多目标物资配置模型。最后，以上海新冠肺炎疫情进行实例分析，结果表明，基于SEIR-LSTM的应急物资需求量预测方法准确率较高，根据分阶段配置模型求出的方案能够满足各个阶段物资分配的要求，验证了提出的模型和算法的有效性。
Research on demand prediction and allocation of emergency medical supplies under major infectious disease epidemics
Scientific and reasonable prediction and allocation of emergency medical supplies is key to improving the prevention and control efficiency of major infectious disease epidemics.Firstly,according to the characteristics of different stages of epidemic evolution,the infectious disease epidemic is divided into six evolutionary scenarios based on four dimensions:key events,medical supplies,epidemic dynamics,and prevention and control levels.Based on the characteristics of epidemic data,an integrated SEIR-LSTM model combining Susceptible Exposed Infectious Recovered(SEIR)and Long Short-Term Memory(LSTM)is proposed to predict the demand for emergency medical supplies at each demand point.This method harnesses LSTM's robust learning ability on time series data to fit the parameters of susceptible and infected individuals in historical epidemic data to better learn the trend of infection rate changes,which is then input into the SEIR model to improve the accuracy of infection number prediction.Then,phased multi-objective material allocation models are formulated based on the characteristics of key evolution stages of infectious disease epidemics.During the regional transmission under the population mobility stage(S3),considerations mainly revolve around economic efficiency and fairness in material distribution.Conversely,during the outbreak under the controlled transmission stage(S4),the urgency of material needs in various epidemic areas is mainly considered,and priority is given to minimizing material shortage losses while ensuring fairness and timeliness of allocation.The multi-objective configuration models are solved using the linear weighting method,with weights adjusted according to the significance of each objective at different stages.Finally,a case study is conducted on the COVID-19 outbreak data in Shanghai.The results show that the proposed SEIR-LSTM prediction model exhibits superior accuracy and faster convergence speed compared to the SEIR model and LSTM model.Furthermore,the solutions derived from the phased allocation model effectively meet all material distribution requirements at each stage,thereby providing robust validation for the efficiency of our proposed model and methodology.

public safetymajor infectious disease epidemicdemand forecastingallocation of emergency medical suppliesepidemic model SEIR(Susceptible Exposed Infectious Recovered)Long Short-Term Memory(LSTM)

袁瑞萍、杨阳、王晓林、多靖赟、李俊韬

展开 >

北京物资学院信息学院,北京 101149

智能物流系统北京市重点实验室,北京 101149

公共安全 重大传染病疫情 需求预测 应急物资配置 传染病模型SEIR 长短期记忆(LSTM)

国家自然科学基金项目北京市教委科技计划重点项目通州区优秀科技创新团队项目

72101033KZ202210037046CXTD2023010

2024

10.13637/j.issn.1009-6094.2023.2448

安全与环境学报
北京理工大学 中国环境科学学会 中国职业安全健康协会

安全与环境学报

CSTPCD北大核心
影响因子：0.943
ISSN：1009-6094
年,卷(期)：2024.24(8)