Fatigue Reliability Based Optimal Replacement Decision Based on First Order Reliability Method

This paper introduces a maintenance decision-making strategy in the general area of the replacement and reliability of mechanical components. The decision-making strategy involves the optimization of the replacement interval based on fatigue failure of mechanical components. Fatigue reliabilities of a component under random cyclic loading need to be evaluated to determine the optimal replacement intervals under fatigue failure. This task is undertaken by determining a reliability factor using an inverse reliability analysis. A reliability-defined ε-N curve (R-ε-N_f curve) can be generated for an empirical ε-N relationship and a “unique” reliability factor by modifying the nominal ε-N_f curve using reliability factors for an assigned reliability. A family of R-ε-N_f curves, which includes the conventional ε-N_f curve, can then be obtained. Hence, fatigue life under specified reliability or reliability based on mission life can be predicted using these curves. This method can efficiently determine replacement intervals for components whose operating costs increase with use and replacement intervals for components subject to failure induced by random process. An example is presented to demonstrate the application of the present method.