Theoretical description of methodology in PHASER (Probabilistic hybrid analytical system evaluation routine)
Probabilistic safety analyses (PSAs) frequently depend on fault tree and event tree models, using probabilities of `events` for inputs. Uncertainty or variability is sometimes included by assuming that the input probabilities vary independently and according to an assumed stochastic probability distribution modes. Evidence is accumulating that this methodology does not apply well to some situations, most significantly when the inputs contain a degree of subjectivity or are dependent. This report documents the current status of an investigation into methods for effectively incorporating subjectivity and dependence in PSAs and into the possibility of incorporating inputs that are partly subjective and partly stochastic. One important byproduct of this investigation was a computer routine that combines conventional PSA techniques with newly developed subjective techniques in a `hybrid` (subjective and conventional PSA) program. This program (PHASER) and a user`s manual are now available for beta use.
- Research Organization:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- AC04-94AL85000
- OSTI ID:
- 199089
- Report Number(s):
- SAND-96-0022; ON: DE96006072
- Resource Relation:
- Other Information: PBD: Jan 1996
- Country of Publication:
- United States
- Language:
- English
Similar Records
PHASER user`s manual version 2.10
Dynamic Probabilistic Safety Assessment Studies for Advanced Reactor Using RAVEN