一种用于管理COVID-19疫情期间不确定群体决策问题中个体不合作行为的共识模型
COVID-19疫情背景下,许多涉及不确定信息的紧急决策问题出现,需要多名个体参与解决,例如疫情期间选择海外航班进入中国的首个入境点。大规模群体决策中,由于专家背景不同,不合作行为常见,对决策效率(决策时间和成本)产生负面影响。既有大规模群体决策方法很少考虑专家不合作行为,本研究提出一种新的共识模型来管理此类行为。模型引入同时考虑模糊偏好值和合作度的群体一致性指标来检测不合作行为,在聚类专家时结合合作度与模糊偏好相似度,并对不合作专家实施动态权重惩罚机制以提升群体共识水平。以COVID-19疫情期间从多伦多飞往北京航班首个入境点选择为例,说明了所提模型的有效性。
要点
- 针对大规模群体决策中专家不合作行为常见且影响决策时间与成本的问题,提出共识模型进行管理。
- 引入同时考虑模糊偏好值和合作度的群体一致性指标,用于检测专家不合作行为。
- 聚类专家时结合合作度与模糊偏好相似度。
- 对不合作专家实施动态权重惩罚机制,以提高群体共识水平。
- 以COVID-19疫情期间从多伦多飞往北京航班首个入境点选择为例说明模型有效性。
适用人群与场景
- 参与COVID-19疫情背景下不确定信息紧急决策问题的大规模群体决策专家,专家背景不同。
诊断标准与关键阈值
- 采用同时考虑模糊偏好值和合作度的群体一致性指标检测专家不合作行为。
处置流程与药物剂量
- 聚类专家时结合合作度与模糊偏好相似度。
- 对不合作专家实施动态权重惩罚机制,以降低低合作度专家对决策质量的负面影响并提高群体共识水平。
研究设计与关键结果
- 研究提出一种新的共识模型,用于管理大规模群体决策问题中专家不合作行为;模型引入同时考虑模糊偏好值和合作度的群体一致性指标检测不合作行为,聚类时结合合作度与模糊偏好相似度,并对不合作专家实施动态权重惩罚机制以提高群体共识水平;以COVID-19疫情期间从多伦多飞往北京航班首个入境点选择的示例说明模型有效性。
证据类型、等级与局限
- 证据类型:方法学研究(共识模型构建与示例说明)。
- 局限:原文未报告局限性
对床旁实践的改变
- 原文未据此给出推荐。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
The COVID-19 pandemic has brought lots of losses to the global economy. Within the context of COVID-19 outbreak, many emergency decision-making problems with uncertain information arose and a number of individuals were involved to solve such complicated problems. For instance, the selection of the first entry point to China is important for oversea flights during the epidemic outbreak given that reducing imported virus from abroad becomes the top priority of China since China has achieved remarkable achievements regarding the epidemic control. In such a large-scale group decision making problem, the non-cooperative behaviors of experts are common due to the different backgrounds of the experts. The non-cooperative behaviors of experts have a negative impact on the efficiency of a decision-making process in terms of decision time and cost. Given that the non-cooperative behaviors of experts were rarely considered in existing large-scale group decision making methods, this study aims to propose a novel consensus model to manage the non-cooperative behaviors of experts in large-scale group decision making problems. A group consistency index simultaneously considering fuzzy preference values and cooperation degrees is introduced to detect the non-cooperative behaviors of experts. We combine the cooperation degrees and fuzzy preference similarities of experts when clustering experts. To reduce the negative influence of the experts with low degrees of cooperation on the quality of a decision-making process, we implement a dynamic weight punishment mechanism to non-cooperative experts so as to improve the consensus level of a group. An illustrative example about the selection of the first point of entry for the flights entering Beijing from Toronto during the COVID-19 outbreak is presented to show the validity of the proposed model.
参考来源
- Li X, Liao H, Wen Z, et al. A consensus model to manage the non-cooperative behaviors of individuals in uncertain group decision making problems during the COVID-19 outbreak. Appl Soft Comput. 2020. PMID: 33519328. doi:10.1016/j.asoc.2020.106879
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。