Reliability Analysis for Dependent Competing Failure System Considering Degradation Rate and Hard Failure Threshold Changes

The failure of many systems often results from competing soft and hard failures, as they experience performance degradation and simultaneous multiple random shocks resulted from the changeable operating environment. On account of the assumption that the performance degradation process follows a linear degradation path model and the shock process follows an extreme shock model, the effects of multiple random shocks on the system performance degradation level, degradation rate, and the hard failure threshold are analyzed. Thereby, under the multiple random shocks, this study propose an innovative reliability model for dependent competing failure systems, which considers that the hard failure threshold will change as well as the degradation rate. Multiple random shocks accelerate the degradation process, causing abrupt increases in performance degradation levels and a rapid acceleration in the degradation rate. Furthermore, in the extreme shock model, previous shocks will cause a decrease of hard failure threshold. Based on the system description and the model assumptions, the expression of system reliability model is established through analysis and deduction, and the calculation is performed through Monte Carlo method. Ultimately, by the numerical experiments of Micro-engine mechanical system and sensitivity analysis of the model's basic parameters, it's proved that the proposed model is effective and accurate.