Modeling dynamic environment effects on dependent failure processes with varying failure thresholds

Abstract

Devices usually fail due to multiple dependent competing failure processes resulting from internal degradation and random shocks, whose behavior may vary in different environments. This paper focuses on systems suffering from randomly occurring shocks described by Poisson processes and internal degradation characterized by linear path models simultaneously, where the wear rate, shock arrival rate, shock load size, and shock damage amount are modulated by Markovian environments, especially the hard failure threshold differs in distinct environments. Reliability analysis is performed where analytical formulas and simulation algorithms for computing reliability indexes of systems are provided, such as the reliability function. Two maintenance models are developed for systems, including a real-time maintenance policy where the computation formula for the availability function is derived, and a periodic inspection policy in which an optimization model is proposed to find the optimal inspection interval that minimizes the average long-run cost rate. Finally, an illustrative example of floating offshore wind turbine systems is given to demonstrate possible applications of the developed models and proposed methods.