Office of Nuclear Regulatory Research
Division of Risk Analysis
As with many subjects, when you’re talking about Probabilistic Risk Assessment (PRA) terms, it can get confusing when the same terms mean different things. To reduce this confusion, the NRC’s Office of Nuclear Regulatory Research decided to produce a glossary of risk-related terms.
In November 2013, we published NUREG-2122, “Glossary of Risk-Related Terms in Support of Risk-Informed Decisionmaking.” This glossary provides a single source for definitions of terms used in risk-informed activities. This document clarifies the terms and provides the proper context for using them. For example, the glossary defines the word “assumption” as “a decision or judgment that is made in the development of a model or analysis.” It then describes how “assumption” is used within the context of a PRA.
We developed the list of terms by first reviewing documents used to support risk-informed activities. Both experienced and new staff contributed to this initial list to ensure terms were not excluded too early. We then screened the list to develop the final glossary by considering each term’s significance to risk-informed decisionmaking.
After we finalized the list, we identified definitions that already existed for these terms. Where a definition already existed, we left it as is. Also, when a definition was consistent among the various sources, it was used as the basis for the final definition in the glossary.
To help in understanding the definitions we used, we included cross-references to other terms in the glossary. Also, for the more complex terms, we included plain language background information to clarify the definition used by experts. This supporting information eliminated technical jargon and — we hope — makes the terms more accessible to non-technical readers.
For example, the glossary goes into quite a bit of detail about “Bayesian analysis,” an uncommon term for the lay audience. The glossary explains that the term applies to a type of analysis used when occurrences of an event are so rare or nonexistent that data from actual event occurrences (frequentist approach) are not reliable.
In addition to defining the terms, the glossary describes how they are used in risk-informed decisionmaking and offers some insights on their history. Where multiple definitions exist, we included them in the descriptions along with an explanation of their differences.
This is our attempt to make what we do a bit less opaque to the non-scientist. We hope it reduces ambiguity and promotes common understanding for better communication. If you have thoughts about future adjustments, feel free to let us know in the comments.