Exploring dynamic Bayesian Belief Networks for intelligent fault management systems

Systems that are subject to uncertainty in their behaviour are often modelled by Bayesian Belief Networks (BBNs). These are probabilistic models of the system in which the independence relations between the variables of interest are represented explicitly. A directed graph is used, in which two nodes are connected by an edge if one is a 'direct cause' of the other. However the Bayesian paradigm does not provide any direct means for modelling dynamic systems. There has been a considerable amount of research effort in recent years to address this. In this paper, we review these approaches and propose a new dynamic extension to the BBN. Our discussion then focuses on fault management of complex telecommunications and how the dynamic bayesian models can assist in the prediction of faults.