Markov Decision Process Models for Optimizing Vaccine Administration

用于优化疫苗管理的马尔可夫决策过程模型

• 批准号: 1131172
• 负责人: Lisa Maillart
• 金额: $ 32.5万
• 依托单位: University of Pittsburgh
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-09-01 至 2015-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1131172&HistoricalAwards=false
• 关键词: Markov; Decision; Process; Models; Optimizing

The research objective of this award is to formulate and analyze Markov decision process models to enable vaccine clinicians to optimally balance the tradeoffs inherent to vaccine administration, especially in remote locations, e.g., rural areas and developing countries. The main tradeoff is one between vaccine "wastage" and missed opportunities for vaccine administration. Wastage is especially acute for vaccines that are packaged in large, multi-dose vials which once opened, must be used within a matter of hours; hence clinicians face difficult tradeoffs between opening a vial to satisfy a potentially small immediate demand, versus retaining the vial to satisfy a potentially large future demand. Furthermore, because the power supply in remote areas of developing countries is often unreliable, the development of -passive cold boxes- that maintain temperature over several weeks - provided they are not opened excessively often is being considered; clinicians deciding when to retrieve vials from these cold boxes will face difficult tradeoffs between retrieving vials for potential immediate use, versus not having vaccines available in the face of future demand. Moreover, the design specifications for these cold boxes are open questions that depend heavily on how their use will impact wastage and coverage in the field.Anticipated outcomes of this research include: (i) optimal vaccine administration policies (e.g., vial opening decisions and cold box retrieval decisions) for different situations (e.g., vaccine type(s), vial size(s), demand profile(s)) as a function of many variables (e.g., time of day, vial inventory, history of cold box openings); (ii) dissemination of operational guidelines that can be easily implemented by clinicians in the field; and (iii) insights into various cold box designs in with respect to vaccine coverage and wastage. The results of the research will be disseminated within the US and internationally through the WHO website and potentially enable millions of additional vaccines to be administered annually throughout the world.

该奖项的研究目标是制定和分析马尔可夫决策过程模型，使疫苗临床医生能够最佳地平衡疫苗管理固有的权衡，特别是在偏远地区，例如，农村和发展中国家。 主要的权衡是疫苗“浪费”和错过疫苗接种机会之间的权衡。 对于包装在大的多剂量小瓶中的疫苗，浪费尤其严重，这些小瓶一旦打开，就必须在几个小时内使用;因此，临床医生面临着在打开小瓶以满足潜在的小的立即需求与保留小瓶以满足潜在的大的未来需求之间的困难权衡。 此外，由于发展中国家偏远地区的电力供应往往不可靠，目前正在考虑开发一种被动式冷藏箱，只要不经常打开，这种冷藏箱可保持温度达数周之久;决定何时从这些冷箱中取回小瓶的临床医生将面临在取回小瓶以用于可能的立即使用之间的困难权衡，而不是在未来的需求面前没有疫苗。 此外，这些冷箱的设计规格是开放性问题，这在很大程度上取决于它们的使用将如何影响该领域的浪费和覆盖率。小瓶打开决定和冷箱取回决定）用于不同的情况（例如，疫苗类型、小瓶尺寸、需求概况）作为许多变量的函数（例如，一天中的时间、小瓶库存、冷箱打开的历史）;（ii）传播可由现场临床医生容易地执行的操作指南;以及（iii）深入了解关于疫苗覆盖率和浪费的各种冷箱设计。 研究结果将通过世卫组织网站在美国和国际上传播，并有可能使世界各地每年增加数百万种疫苗。