When using vibration-based maintenance, it is easier to detect damage in rolling element bearings. But the damage severity and risk of bearing's failure, if no maintenance action such as replacement is done, can not be assessed accurately. When the vibration level of a rolling element bearing is increasing or it is close to (or greater than) a predetermined replacement level, the assessment of the probability of a bearing’s failure enhances the decision making process. Thus, at each vibration measurement occasion a decision has to be made regarding whether the bearing needs to be maintained and how urgent it actually is.

The paper describes a graphical method for both assessing the probability of failure of damaged rolling element bearing and predicting the residual of its effective life. It uses history of the bearing in question and life statistics as well as the vibration levels at failure/condition-based replacements of past bearings of the same design or of very similar performance and duties. This method is based on the concept of Total Time on Test, TTT-plot. The application of TTT-plot on age and condition-based maintenance techniques are discussed and described by two examples. This graphical method is developed from previous work by the author and others.