中国安全科学学报 ›› 2020, Vol. 30 ›› Issue (7): 19-26.doi: 10.16265/j.cnki.issn1003-3033.2020.07.004

• 安全社会科学与安全管理 • 上一篇    下一篇

基于DBN的疫苗运输质量安全风险监测方法

谢小良1,2 教授, 褚琦**1,2, 张淑君3 主任医师, 卫国4 教授, 成佳祺1,2   

  1. 1 湖南工商大学 数学与统计学院,湖南 长沙 410205;
    2 统计学习与智能计算湖南省重点实验室,湖南 长沙 410205;
    3 湖南省疾病预防控制中心,湖南 长沙 410005;
    4 北卡罗来纳州立大学数学与计算机科学系,美国 北卡罗来纳 彭布罗克 28372
  • 收稿日期:2020-04-10 修回日期:2020-06-08 出版日期:2020-07-28 发布日期:2021-07-15
  • 通讯作者: ** 褚 琦(1998—),女,浙江金华人,硕士研究生,主要研究方向为统计分析与智能决策。E-mail:15967183607@163.com。
  • 作者简介:谢小良 (1964—),男,湖南长沙人,博士,教授,硕士生导师,主要从事数据分析与智能决策方面的研究。E-mail: Hnucmath207@163.com。
  • 基金资助:
    国家社会科学基金资助(19BTJ011)。

DBN-based monitoring method of vaccine transportation quality and safety risk

XIE Xiaoliang1,2, CHU Qi1,2, ZHANG Shujun3, WEI Guo4, CHENG Jiaqi1,2   

  1. 1 School of Mathematics and Statistics, Hunan University of Technology and Business, Changsha Hunan 410205, China;
    2 Key Laboratory of Hunan Province for Statistical Learning and Intelligent Computation, Changsha Hunan 410205, China;
    3 Hunan Province Center for Discase Control and Prevention, Changsha Hunan 410005, China;
    4 Department of Mathematics & Computer Science, University of North Carolina at Pembroke, Pembroke 28372, USA
  • Received:2020-04-10 Revised:2020-06-08 Online:2020-07-28 Published:2021-07-15

摘要: 为识别、度量和应对疫苗运输过程中冷链控制和流向追溯2大风险,将疫苗运输全过程描述为一个复杂动态系统,通过分析疫苗运输质量安全风险影响因素,构建以疫苗采购、储存、运输、装卸为风险链的评价指标体系,建立基于动态贝叶斯网络(DBN)的疫苗运输质量安全风险评估(TQSRA)模型;选取2009年以来OpenLaw网上公布的328例疫苗安全事件为案例,搜索相关数据,借助GeNIe 2.0 可视化软件,实证检验TQSRA模型的效度及敏感性,探索疫苗运输质量安全风险的敏感点,明确防控重点。研究结果表明:疫苗来源、疫苗从业人员配备和培训、疫苗运输信息的全程记录与检测,是疫苗运输质量安全监管的主要因素;实例证明TQSRA模型性能稳定、实用、有效。

关键词: 疫苗运输, 动态贝叶斯网络(DBN), 灵敏度分析, 运输质量安全风险评估(TQSRA), 监测

Abstract: In order to identify, measure and respond to two major risks of cold chain control and flow tracing, vaccine transportation process was modeled as a complex dynamic system, and by analyzing influencing factors of vaccine quality and safety risks during transportation, an evaluation index system was constructed in which vaccine procurement, storage, distribution, and loading were determined as key elements in risk chain. Secondly, a vaccine TQSRA model was developed based on DBN. 328 cases of vaccine safety incidents since 2009 were selected from OpenLaw website for statistical data learning by using GeNIe 2.0 visualization software, and an empirical study was conducted to examine validity and sensitivity of proposed TQSRA model. Finally, critical sensitive linkages of quality and safety risks during transportation were explored, and focuses of prevention and control were clarified. The results show that quality of cold chain transport vehicles and storage facilities, comprehensive quality of vaccine practitioners, and entire record and inspection of transport information are main factors affecting vaccine transportation quality and safety supervision. And TQSRA model is proved to be a stable, practical and effective assessment method.

Key words: vaccine transportation, dynamic Bayesian network(DBN), sensitivity analysis, transportation quality and safety risk assessment(TQSRA), monitoring

中图分类号: