Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Risk analysis using a hybrid Bayesian-approximate reasoning methodology.

Conference ·

Analysts are sometimes asked to make frequency estimates for specific accidents in which the accident frequency is determined primarily by safety controls. Under these conditions, frequency estimates use considerable expert belief in determining how the controls affect the accident frequency. To evaluate and document beliefs about control effectiveness, we have modified a traditional Bayesian approach by using approximate reasoning (AR) to develop prior distributions. Our method produces accident frequency estimates that separately express the probabilistic results produced in Bayesian analysis and possibilistic results that reflect uncertainty about the prior estimates. Based on our experience using traditional methods, we feel that the AR approach better documents beliefs about the effectiveness of controls than if the beliefs are buried in Bayesian prior distributions. We have performed numerous expert elicitations in which probabilistic information was sought from subject matter experts not trained In probability. We find it rnuch easier to elicit the linguistic variables and fuzzy set membership values used in AR than to obtain the probability distributions used in prior distributions directly from these experts because it better captures their beliefs and better expresses their uncertainties.

Research Organization:
Los Alamos National Laboratory
Sponsoring Organization:
DOE
OSTI ID:
975658
Report Number(s):
LA-UR-01-4153
Country of Publication:
United States
Language:
English

Similar Records

A comparison of approximate reasoning results using information uncertainty
Conference · Wed Dec 31 23:00:00 EST 2008 · OSTI ID:956519

Development of a Severe Accident Analysis Engine Using Approximate Reasoning
Journal Article · Fri Jul 01 00:00:00 EDT 2016 · Transactions of the American Nuclear Society · OSTI ID:23042723

LinguisticBelief and PoolEvidence
Software · Tue Mar 11 00:00:00 EDT 2008 · OSTI ID:1231199