Rolling element bearing failures in paper mill machines are considered in relation to their critical role in the machine function. The use of expensive, sophisticated and highly automated equipment and machines and the intention to achieve higher quality products, longer machine life, higher machinery effectiveness and safer operating processes were the main driving force motivating efforts to improve maintenance concept during the last 50 years. In this paper, a conceptual model that integrates the available condition information, the deterministic models used in condition monitoring based upon mechanical theory and the probabilistic models used in the area of operational research is developed and its applicability is discussed. This models covers fault detection of a mechanical component such as a rolling element bearing, prediction of its vibration level in the near future, assessment of the probability of failure of a component over a finite period of time of interest.