Modeling condition based maintenance decision support

Condition monitoring is growing in popularity in industry with considerable sums now being spent on condition monitoring hardware and software. This paper focuses upon the prognosis aspect of condition monitoring and associated condition based maintenance decision making. The key element in such decision making is the concept of the conditional residual time, where the residual time of the component depends upon the age but is also influenced by the condition information obtained to date. A recent development in predicting the residual time distribution of a set of six bearings experimented on a test rig is introduced and the associated decision model is discussed.