Evaluation of generalised proportional intensities models with application to the maintenance of gas turbines

In this paper generalised proportional intensities models are fitted to gas turbine generator failure and maintenance data. The systems of interest have identical maintenance procedures and information about preventive maintenance (PM) and corrective maintenance (CM) activities is known. Closed-form maximum log-likelihood estimates of the model parameters are evaluated, and the model goodness-of-fit is measured. A simulation study is conducted to determine an optimal cost effective PM schedule. Our analysis and results facilitate decision support for determining the cost-optimal maintenance strategy.