Process model-based Dynamic Bayesian Networks for Prognostic

Alexandre Muller (1) , Philippe Weber (1) , Abdeljabbar Ben Salem (1, 2)

1 CRAN - Centre de Recherche en Automatique de Nancy
2 INRETS - Institut National de Recherche sur les Transports et leur Sécurité

The degradation checking of critical components is one of the efficient means to minimize the non expected stops of production system and to reduce its costs. In this purpose, a methodology is proposed to design a prognosis process taking into account the behaviour of environmental events (exogenous inputs). The main contribution is the development of the probabilistic model which is based on the translation of a preliminary process model into Dynamic Bayesian Network (DBN). Specifications are given to assure the coherence both of the translation and the resulted model. These requirements are formalized according to the NIAM/ORM method towards its semantic strength, clarity and stability. An application is proposed on a water heater system.

Mots clés: Process approach; cycle problem; bayesian network; prognostic

Domaines: Automatique

Alexandre Muller, Philippe Weber, Abdeljabbar Ben Salem. Process model-based Dynamic Bayesian Networks for Prognostic. Fourth International Conference on Intelligent Systems Design and Applications, ISDA 2004, Aug 2004, Budapest, Hungary. pp.CDROM. ⟨hal-00135818⟩