Zhu, Dongfeng, Wang, Chengdong, Groth, Katrina, Mosleh, Ali A PRA Software Platform for Hybrid Causal Logic Risk Models (Inproceedings) Zio, Enrico; Ho, Vincent; Kao, Tsu-Mu (Ed.): Proceedings of the International Conference on Probabilistic Safety Assessment and Management (PSAM 9), Hong Kong, 2008.

BibTeX

@inproceedings{Zhu2008,
title = {A PRA Software Platform for Hybrid Causal Logic Risk Models},
author = {Dongfeng Zhu and Chengdong Wang and Katrina Groth and Ali Mosleh},
editor = {Enrico Zio and Vincent Ho and Tsu-Mu Kao},
year = {2008},
date = {2008-05-01},
booktitle = {Proceedings of the International Conference on Probabilistic Safety Assessment and Management (PSAM 9)},
address = {Hong Kong},
abstract = {Conventional methods for Probabilistic Risk Assessment (PRA) are based on Boolean logic representing the deterministic relations among various basic events. PRA software tools construct the scenarios using graphical representation of the Boolean relations in the form of Fault Trees, FT, and Event Trees , ET, (as in RiskMan, Risk Spectrum, Saphire, etc.) or Fault Trees and Event Sequence Diagrams, ESDs (as in QRAS). This paper introduces a PRA software platform that combines the power of Event Sequence Diagrams and Fault Trees ideal for modeling engineered systems risks and hazards, with the flexibility of the Boolean logic-based methods Bayesian Belief Network. The three layers modeling approach allows the most appropriate modeling techniques to be applied in different domains of the system. The underlying computational methodology has been implemented in a new PRA software platform – Integrated Risk Information System (IRIS).},
keywords = {aviation, Bayesian Networks, causal models, Hybrid causal logic, IRIS, Probabilistic risk assessment (PRA), software, transportation safety},
pubstate = {published},
tppubtype = {inproceedings}
}

Abstract

Conventional methods for Probabilistic Risk Assessment (PRA) are based on Boolean logic representing the deterministic relations among various basic events. PRA software tools construct the scenarios using graphical representation of the Boolean relations in the form of Fault Trees, FT, and Event Trees , ET, (as in RiskMan, Risk Spectrum, Saphire, etc.) or Fault Trees and Event Sequence Diagrams, ESDs (as in QRAS). This paper introduces a PRA software platform that combines the power of Event Sequence Diagrams and Fault Trees ideal for modeling engineered systems risks and hazards, with the flexibility of the Boolean logic-based methods Bayesian Belief Network. The three layers modeling approach allows the most appropriate modeling techniques to be applied in different domains of the system. The underlying computational methodology has been implemented in a new PRA software platform – Integrated Risk Information System (IRIS).