Bayesian Network Models for Making Maintenance Decisions from Data and Expert Judgment

To maximize asset reliability cost-effectively, maintenance should be scheduled based on the likely deterioration of an asset. A number of types of statistical model have been proposed for predicting this but they have important practical limitations. We present a Bayesian network model that can be used for maintenance decision support to overcome these limitations. The model extends an existing statistical model of asset deterioration, but shows how i) failure data from related groups of asset can be combined, ii) data on the condition of assets available from their periodic inspection can be used iii) expert knowledge of the causes deterioration can be combined with statistical data to adjust predictions and iv) the uncertain effects of maintenance actions can be modelled. We show how the model could be used for a range of decision problems, given typical data likely to be available in practice.