Modelling Non-Deterministic Causal Mechanisms involving Resilience in Risk Analysis
| Autor: | A. De Galizia, Philippe Weber, Carole Duval, E. Serdet, Christophe Simon, Benoît Iung |
|---|---|
| Přispěvatelé: | EDF (EDF), Centre de Recherche en Automatique de Nancy (CRAN), Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), CIFFRE DeGalizia, EDF-CRAN, Simon, Christophe, Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL) |
| Rok vydání: | 2016 |
| Předmět: |
Fault tree analysis
0209 industrial biotechnology Propagation of uncertainty Engineering Sociotechnical system Resilience Risk analysis business.industry 020209 energy Sociotechnical systems Probabilistic logic Bayesian network Statistical model 02 engineering and technology [SPI.AUTO]Engineering Sciences [physics]/Automatic [SPI.AUTO] Engineering Sciences [physics]/Automatic Bayesian networks 020901 industrial engineering & automation Risk analysis (engineering) Canonical models Control and Systems Engineering 0202 electrical engineering electronic engineering information engineering Canonical model Probabilistic modelling business |
| Zdroj: | 8th IFAC Conference on Manufacturing Modelling, Management and Control, MIM 2016 8th IFAC Conference on Manufacturing Modelling, Management and Control, MIM 2016, Jun 2016, Troyes, France |
| ISSN: | 2405-8963 |
| DOI: | 10.1016/j.ifacol.2016.07.625 |
| Popis: | International audience; In risk analysis, deterministic propagation of uncertainty as in fault trees models is not sufficient for sociotechnical systems. In recent years some approaches have been developed as Integrated Risk Analysis (IRA) to address different risks causalities linking human, organizational, technical and environmental factors in a unified framework. This framework relies on a Performance Shaping Factors (PSF) based model and constitutes an alternative approach different from that currently adopted at EDF for PSA (Probabilistic Safety Assessment) purposes. The IRA method is supported by Bayesian Networks (BNs) to model non-deterministic causal patterns among system variables. Thus, this paper aims to investigate on how to consider resilient influences in causal probabilistic modelling. In particular, it is focused on how to consider mitigation mechanisms in quantifying organizational influences on a human activity in sociotechnical systems. It leads to propose a formalism to represent a mitigation mechanism in causal interactions between pathogenic and resilient influences in a probabilistic model. Finally, the feasibility of our proposal is shown on an illustrative case declined in the framework of the IRA approach |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|