Application of physical failure models to enable usage and load based maintenance

The efficiency of a preventive maintenance process largely depends on the ability to predict the replacement intervals of components. Considering the actual usage of the system increases the accuracy of this prediction. The present paper proposes two new maintenance concepts, that combine the benefits of traditional static concepts and condition based maintenance. These new concepts, usage based maintenance and load based maintenance, apply usage or load parameters that are monitored during service to perform a physical model-based assessment of the system condition. The new concepts are positioned within the range of existing maintenance concepts. Also, the role of physical models in maintenance modelling in general is explained and the origin of uncertainty in the predicted service life is discussed. Moreover, it is demonstrated how the monitoring of usage, loads or condition can reduce this uncertainty and increase the service life, by extending existing work in this field. Finally, the different concepts are applied to a gas turbine blade case study to illustrate the benefits of the proposed concepts.