Use of probabilistic relational model (PRM) for dependability analysis of complex systems

Gabriela Medina-Oliva (1) , Philippe Weber (1) , Eric Levrat (1) , Benoît Iung (1)

1 CRAN - Centre de Recherche en Automatique de Nancy

This paper proposes a methodology to develop a aided decision-making tool for assessing the dependability and performances (i.e. reliability) of an industrial system. This tool is built on a model based on a new formalism, called the probabilistic relational model (PRM) which is adapted to deal with large and complex systems. The model is formalized from functional, dysfunctional and informational studies of the technical industrial systems. An application of this methodology is done on a water heater system to illustrate its feasibility and added value.

Mots clés: Probabilistic relational model; maintenance; system performances

Domaines: Modélisation et simulation

Gabriela Medina-Oliva, Philippe Weber, Eric Levrat, Benoît Iung. Use of probabilistic relational model (PRM) for dependability analysis of complex systems. 12th IFAC Symposium on Large Scale Systems: Theory and Applications, LSS 2010, Jul 2010, Villeneuve d'Ascq, France. pp.CDROM. ⟨hal-00549877⟩