Estimating Remaining Useful Lifetime Considering Effects of Different Process Stress on Degradation

The process of health monitoring and correct prognosis of time to failure occurrence is still considered by many researchers of systems reliability to seek more effective use of available facilities. In the present study, a one-component system with four general failure mechanisms has been considered, one of which is excessive degradation under normal operating conditions and the other is accelerating the degradation process. In this paper, continuous degradation of the Gamma process is considered with two discrete and continuous noise factors or stresses, which create three different failure mechanisms in the accelerated lifetime condition. The discrete noise factor follows the Poisson distribution function and the continuous noise factor follows the normal distribution function. These four situations are investigated in the present paper and in each of them, the estimation of the reliability function and the remaining useful lifetime (RUL) or health prognosis of the equipment is obtained in order to reduce the probability of failure in this single-component system. A hybrid approach using statistical process control (SPC) with data transformation method is used to monitor the noise factors. It was also shown that with the significance of the noise factor or factors the scale parameter of the gamma distribution increases and, the reliability and remaining useful lifetime decrease. An example is solved at the end to illustrate the proposed method.