Inconsistent terminology: An RIPB stumbling block

The American Nuclear Society's Risk-informed, Performance-based Principles and Policy Committee (RP3C) held another presentation in its monthly Community of Practice (CoP) series. Former RP3C chair N. Prasad Kadambi opened the June 27 meeting with brief introductory remarks about the RP3C and the recent incorporation of risk-informed, performance-based (RIPB) design principles in the Nuclear Regulatory Commission's Part 53 framework. He then welcomed this month's speaker: Lyndsey Fyffe, a nuclear safety engineer with Strategic Management Solutions, who presented "RIPB Terminology Across the Nuclear Industry."

Some background: RP3C is a special committee created by the ANS Standards Board and currently chaired by Steven Krahn that provides guidance to ANS standards committees on the use of RIPB methods. The CoP is part of RP3C's charter, which includes training and knowledge-sharing of RIPB principles to exchange ideas outside of the normal management and project processes. CoPs are used frequently by organizations to help break down barriers that impede the flow of information.

Framing the study: Fyffe's presentation centered on a comparative study of the varying, inconsistent RIPB-related terminology used by different organizations in the nuclear industry.

She kicked off her talk with a quick thought exercise: Imagine yourself as a new nuclear engineer. You spend two years attending meetings, reading documents, and training before you are confident and competent with the language your team uses. Then, at year five of your career, you're on LinkedIn and you find a perfect new position and decide to apply. But, there's one catch: Your old team was an NRC licensee, and your new team is a Department of Energy contractor. You find yourself-despite being in similar a role-having to spend the next two years learning a whole new language.

The consequences of this disharmony are numerous. It increases the chance of miscommunication, even among RIPB subject matter experts; it weakens the industry's ability to collaborate between both national and international stakeholders; it unduly burdens those making career changes; it throttles workforce development; it complicates regulatory requirements; and it increases the difficulty of communicating with the public.

Some details and results: Fyffe looked at three U.S.-based guidance and regulatory documents for nonreactor nuclear facilities, one from the DOE, one from the NRC, and one from the Nuclear Energy Institute. She limited the scope of the study to a desktop exercise-comparing language rather than interviewing stakeholders.

While many areas of terminology were found to have some level of divergence, definitions for frequency and control in particular showed a large degree of disharmony.

As depicted in the above chart, not only do NRC, DOE, and NEI use different terms for similar ranges, their ranges also vary significantly, differ in count, and scope. Fyffe concludes, "it would be pretty difficult for us to be communicating and making sure that we're all saying the same thing. You would almost need a decoder ring to figure out what people are talking about."

How to help: Fyffe argues that harmonizing terminology will reduce ambiguity and improve clarity in risk assessments, as well as improve the ability to verify and validate risk analysis models and improve communication in risk debates. That work will come with significant barriers, however. The resources required to retrain staff, questions over whose versions of terms will be adopted, and the general difficulty of changing legislation all present real challenges.

One small way to get started is updating ANS's document endorsing the use of RIPB methods: Incorporating Risk-Informed and Performance-Based Approaches/Attributes in ANS Standards. To begin that work, Fyffe is calling for volunteers who have used the document to get their firsthand perspectives on what works well in it and what could use clarification. To learn more about how to help, watch the full webinar here.