| Abstract | Over the life time of any piece of complex equipment, the likelihood of a failure and the cost of its repair will change. The best machine learning classifier, for predicting failures, is dependent on these values. This paper presents a way of visualizing expected cost which gives a clear picture as to when a particular classifier is the right one to use. Of equal importance, it also shows when a classifier should not be used and a more traditional maintenance policy is the better choice. It distinguishes the conditions when it is best to wait until a part breaks before taking action, or when it is best replace it routinely, at regular intervals. This paper demonstrates how overall, this visualization method gives maintenance personnel the means to adapt to changing failure rates and changing costs. |
|---|
