Probability Elicitation with Groups of Experts

专家组的概率推导

• 批准号: 9730704
• 负责人: Detlof von Winterfeldt
• 金额: $ 22.87万
• 依托单位: University of Southern California
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1998
• 资助国家: 美国
• 起止时间: 1998-06-01 至 2002-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9730704&HistoricalAwards=false
• 关键词: Probability; Elicitation; Groups; Experts

Risk analysis is a formal method to assess the probabilities and consequences of hazards to human health and the environment. The assessment of probabilities and consequences is complicated, when little data exist, when models are inadequate, and when experts disagree. In these situations, risk analysts increasingly use formal processes for eliciting probabilities from groups of experts. The intention is to incorporate a broad spectrum of scientific opinions and to acknowledge and incorporate scientific uncertainties as well as scientific disagreements. This project will study how to best involve scientific experts with different approaches and different opinions in this process. The study will compare several methods of eliciting probabilities and consequences of risks from groups of experts, including eliciting them separately versus in groups and allowing disagreements versus forcing consensus. The study will be conducted by performing experiments both with engineering students and with experts in the field. The results should contribute substantially to improve the science and practice of risk assessment.

风险分析是一种评估危害人类健康和环境的可能性和后果的正式方法。 当数据很少、模型不充分以及专家意见不一致时，对概率和后果的评估是复杂的。 在这些情况下，风险分析师越来越多地使用正式的过程来从专家组中得出概率。 其目的是纳入广泛的科学意见，并承认和纳入科学上的不确定性以及科学上的分歧。 该项目将研究如何最好地让不同方法和不同意见的科学专家参与这一进程。 这项研究将比较几种从专家组中得出风险概率和后果的方法，包括分别与分组得出风险概率和后果的方法，以及允许分歧与强迫达成共识的方法。 这项研究将通过与工程专业的学生和该领域的专家进行实验来进行。 研究结果将大大有助于改进风险评估的科学和实践。