作者: Roger Flage , Terje Aven
DOI:
关键词: Component (UML) 、 Mathematics 、 Risk analysis 、 Order (exchange) 、 Econometrics 、 Relation (database) 、 Decision support system 、 Uncertainty analysis 、 Sensitivity analysis 、 Risk analysis (engineering) 、 Expected value
摘要: A quantitative risk analysis (QRA) should provide a broad, informative and balanced picture of risk, in order to support decisions. To achieve this, proper treatment uncertainty is prerequisite. Most approaches QRA seem be based on the thinking that relates calculated probabilities expected values. This causes difficulties when it comes communicating what results mean, could easily lead weakened conclusions if large uncertainties are involved. An alternative approach hold uncertainty, not probability, as main component regard purely epistemic-based expressions uncertainty. In paper latter view taken, we describe components description following this approach. We also indicate how decision-making. important issue addressed communicate shortcomings limitations Sensitivity plays key role regard. Examples included illustrate ideas findings.