Guidance on the Treatment of Uncertainties Associated with PRAs in Risk-Informed Decision Making - Draft Report For Comment (NUREG-1855)
This NUREG-series publication was issued for public comment, and the comment period is now closed.
On this page:
Download complete document
Manuscript Completed: November 2007
Date Published: November 2007
M. Drouin (NRC Project Manager)1, G. Parry2
J. Lehner3, G. Martinez-Guridi3
J. LaChance4, T. Wheeler4
3Brookhaven National Laboratory
Upton, NY 11973
4Sandia National Laboratories
Albuquerque, NM 87185
1Office of Nuclear Regulatory Research
2Office of Nuclear Reactor Regulation
U.S. Nuclear Regulatory Commission
Washington, DC 20555-0001
This document provides guidance on how to treat uncertainties associated with probabilistic risk assessment (PRA) in risk-informed decision making. The objectives of this guidance include fostering an understanding of the uncertainties associated with PRA, the impact of the uncertainties on the results of the PRA, and the uncertainties in the context of the decision making. The guidance in this document focuses on the use of PRA insights and results and ways to address the associated uncertainties. Consequently, the scope of the guidance contained in this report is limited to addressing the uncertainties associated with the use of the results of risk models.
In implementing risk-informed decision making, the US Nuclear Regulatory Commission expects that appropriate consideration of uncertainty will be given in analyses and interpretation of findings. Such consideration should include using a program of monitoring, feedback, and corrective action to address significant uncertainties. To meet this objective, it is necessary to understand the role that PRA results play in the context of the decision process. Defining the context includes providing an overview of the risk-informed decision making process itself.
With the context defined, the characteristics of a risk model and, in particular, a PRA need to be understood. This understanding includes a recognition of the different forms of uncertainty which include aleatory and epistemic. A PRA, as a probabilistic model already characterizes aleatory uncertainty. The focus of this document is epistemic uncertainty. Therefore, guidance is given on identifying and describing the different types of sources of epistemic uncertainty including the different ways that they are treated. The different types of epistemic uncertainty include parameter, model, and completeness uncertainties.
The final part of the guidance includes addressing the uncertainty in PRA results in the context of risk-informed decision making and, in particular, the interpretation of the results of the uncertainty analysis when comparing PRA results with the acceptance criteria established for a specified application. In addition, guidance is provided for addressing the other elements contributing to completeness uncertainty in risk-informed decision making (e.g., unknown phenomena that have not been recognized or factors that have been identified but for which there is no agreed on method for addressing them in PRAs).