Reliability data assessment may be a part of integrated regional risk assessment and safety management (IRRASM).
This article considers the description and modelling of technical events within reliability analysis.
Normally, problems arise in a practical approach, for example in the description of components and their failures, to warrant homogeneity of samples and to choose an adequate source of data.
Beside statistical modelling, an analysis has to avoid the pitfall rushing to conclusions : similar-looking failure data are seductive, but their usage of different data sources needs much additional information about methods, models and descriptions of components.
Some of these difficulties are pointed out by examples.
Mots-clés Pascal : Théorie, Théorie fiabilité, Système grande taille, Risque accidentel, Prévention accident, Gestion information, Maniement donnée, Modèle mathématique, Méthode statistique, Analyse dommage, Evaluation risque
Mots-clés Pascal anglais : Failure rate, Integrated regional risk assessment, Generic data, Plant specific data, Process industry, Theory, Reliability theory, Large scale systems, Hazards, Accident prevention, Information management, Data handling, Mathematical models, Statistical methods, Failure analysis, Risk assessment
Notice produite par :
Inist-CNRS - Institut de l'Information Scientifique et Technique
Cote : 96-0453697
Code Inist : 002B30B. Création : 10/04/1997.