Statistically based uncertainty assessments in nuclear risk analysis

Conference · Thu Jan 01 04:00:00 EST 1987

OSTI ID:6258200

Spencer, F W; Diegert, K V; Easterling, R G

Over the last decade, the problems of estimation and uncertainty assessment in probabilistics risk assessment (PRAs) have been addressed in a variety of NRC and industry-sponsored projects. These problems have received attention because of a recognition that major uncertainties in risk estimation exist, which can be reduced by collecting more and better data and other information, and because of a recognition that better methods for assessing these uncertainties are needed. In particular, a clear understanding of the nature and magnitude of various sources of uncertainty is needed to facilitate descision-making on possible plant changes and research options. Recent PRAs have employed methods of probability propagation, sometimes involving the use of Bayes Theorem, and intended to formalize the use of ''engineering judgment'' or ''expert opinion.'' All sources, or feelings, of uncertainty are expressed probabilistically, so that uncertainty analysis becomes simply a matter of probability propagation. Alternatives to forcing a probabilistic framework at all stages of a PRA are a major concern in this paper, however.

• Research Organization: Sandia National Labs., Albuquerque, NM (USA)
• DOE Contract Number: AC04-76DP00789
• OSTI ID: 6258200
• Report Number(s): SAND-87-0819C; CONF-870898-2; ON: DE87013724
• Country of Publication: United States
• Language: English

Related Subjects

22 GENERAL STUDIES OF NUCLEAR REACTORS
220900* -- Nuclear Reactor Technology-- Reactor Safety
DATA COVARIANCES
MATHEMATICAL MODELS
MECHANICS
NUCLEAR FACILITIES
NUCLEAR POWER PLANTS
POWER PLANTS
PROBABILISTIC ESTIMATION
PROBABILITY
REACTOR SAFETY
RISK ASSESSMENT
SAFETY
STATISTICAL MECHANICS
STATISTICAL MODELS
THERMAL POWER PLANTS