Historical Published Maintenance Data: What Can It Tell Us About Reliability Modelling?

The paper discusses physical assumptions versus statistical assumptions illustrated by some reliability data, which if modelled ignoring physical assumptions (which are more important) gives incorrect results. The importance of physical assumptions can be extended to each stage of systems and service to improve the modelling being done (or that should be done). Explanatory physical information regarding systems is available within reliability data. It just has to be searched for. Four typical data sets are analysed in detail in the paper. In these analyses, an attempt has been made to convert operational time to ‘failure’ to calendar time to maintenance event. There are many other data sets that this straightforward methodology can be applied to. Previous research has developed methods for analysing the operational time of individual equipment but a more comprehensive approach is required to model the whole maintenance system taking into account both the planned and unplanned events.