03-4624. Handbook of Parameter Estimation for Probabilistic Risk Assessment (PRA)

AGENCY:

Nuclear Regulatory Commission (NRC).

ACTION:

Announcement of issuance for public comment, availability, and public meeting.

SUMMARY:

The U.S. Nuclear Regulatory Commission (NRC) has issued for public comment a document entitled “Handbook of Parameter Estimation for Probabilistic Risk Assessment.” This handbook provides guidance on sources of information and methods of estimating distributions of parameters used in probabilistic risk assessment. This includes determination of both plant-specific and generic estimates for initiating event frequencies, component failure rates and unavailability, and equipment non-recovery probabilities. Interested individuals may obtain a copy of this document from the person identified under the caption FOR FURTHER INFORMATION CONTACT.

DATES:

Please submit comments by May 23, 2003. Comments received after this date will be considered if is practical to do so, but the NRC Staff is able to ensure consideration only for comments received on or before this date.

A public meeting will be held on April 2, 2003, from 8:30 a.m. until 4:30 p.m. The purpose of this meeting is to discuss the document and give guidance for its review and to address external stakeholder comments received before the public meeting. The primary authors will be available at the public meeting to answer questions and discuss items of interest to the public.

The NRC is interested in receiving comments on specific topics reviewers wish to discuss at the public meeting. Reviewers are requested to contact Mr. Arthur Salomon at 301-415-6552 or e-mail [e-mail] by March 21, 2003, with comments or issues they wish to have addressed at the public meeting.

ADDRESSES:

Please submit comments to Chief, Rules and Directives Branch, Office of Administration, U.S. Nuclear Regulatory Commission, Washington, DC 20555-0001. Or, you may deliver comments to 11545 Rockville Pike, Rockville, Maryland, between 7:30 a.m. and 4:30 p.m. Federal workdays.

The public meeting will be held at Two White Flint North, Room T10 A1, 11545 Rockville Pike, Rockville, Maryland.

The draft document and certain other documents related to this action may be examined in the NRC Public Document Room, 11555 Rockville Pike, Public File Area O1 F21, Rockville, Maryland.

FOR FURTHER INFORMATION CONTACT:

Arthur D. Salomon, Division of Risk Analysis and Applications, Office of Nuclear Regulatory Research, U.S. Nuclear Regulatory Commission, Washington, DC 20555-0001. Telephone: [phone], e-mail: [e-mail].

SUPPLEMENTARY INFORMATION:

During the last several years the NRC and the nuclear industry have recognized that PRA has evolved to the point where it can be used as a tool in the regulatory decision-making process. The increased use of PRA has led to the conclusion that the PRA scope and model must be commensurate with the applications. Several procedural guides and standards (such as the “Standard For Probabilistic Risk Assessment For Nuclear Power Plant Applications” published by The American Society of Mechanical Engineers in 2002) have been and are being developed that identify requirements for the PRA models. This handbook was prepared to supplement these documents. It provides a compendium of good practices that a PRA analyst can use to develop parameter distributions required for quantifying PRA models. It complements the ASME PRA Standard by providing guidance for estimation of parameters related to initiating events; by describing statistical techniques for estimating parameters used in analysis of basic events such as equipment failures; and by addressing methods for estimating equipment unavailability. Furthermore, the handbook includes appendices describing the basics of probability and statistics used in performing the more detailed analyses discussed in the main portion of the document.

The handbook provides the basic information needed to generate estimates of the parameters commonly used in PRA analysis. It begins by describing the probability models and plant data used to evaluate each of the parameters. Possible sources for the plant data are identified and guidance on the collection, screening, and interpretation is provided. The statistical techniques (both Bayesian and classical methods) required to analyze the collected data and test the validity of statistical models are described. Examples are provided to help the PRA analyst utilize the different techniques. The handbook also provides advanced techniques that address modeling of time trends. Methods for combining data from a number of similar, but not identical sources are also provided. This includes empirical and hierarchical Bayesian approaches. Examples are provided to guide the analyst. The handbook does not provide guidance on parameter estimation for all of the events included in a PRA. Specifically, common cause failure and human error probabilities are not addressed. In addition, guidance is not provided regarding the use of expert elicitation. However, suggested references regarding these technical areas are provided in the handbook introduction.

Dated at Rockville, Maryland, this 21st day of February, 2003.

For the Nuclear Regulatory Commission.

Scott F. Newberry,

Director, Division of Risk Analysis and Applications, Office of Nuclear Regulatory Research.