PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

A generic approach to model complex system reliability using graphical duration models
R. Donat, L. Bouillaut, P. Aknin, Philippe Leray and D. Levy
In: Mathematical Methods in Reliability: Methodology and Practice (MMR 2007),, 1-4 july 2007, Glasgow, UK.


Nowadays, reliability analysis has become an integral part of system design and operating. This is especially true for systems performing critical tasks. Moreover, recent works in reliability involving the use of probabilistic graphical models, also known as bayesian networks, have been proved relevant. This paper aims to describe a general methodology to model the stochastic degradation process of a complex system, allowing any kind of state sojourn distributions along with an accurate context description. We meet these objectives using a specific dynamic graphical model, namely a graphical duration model. In this article, we give qualitative and quantitative descriptions of the proposed model and describe a simple algorithm to estimate the system reliability and some of its related metrics. Finally, we illustrate this approach by applying our methodology to a three-states system subjected to one context variable and with non exponential duration distributions.

EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
ID Code:3768
Deposited By:Philippe Leray
Deposited On:21 February 2008