Abstract:

Aiming at the problem of longlifetime reliability assessment in the assumption of unknown life distribution, a new method on reliability assessment for degradation data is proposed, which is different from the traditional method based on the degradation path and degradation measure distribution. Time is treated as a similaraccelerated factor and used as a covariate. The relationship between the lifetime and the degradation failure rate of the product is described by the proportional hazard model and the proportional rate model is built. The degradation distribution is determined through the divergence of the distribution set， and the degradation basic failure rate function and its proportional rate coefficient are obtained. The linear regression model is used to evaluate the parameters of the similaraccelerated model， and the degradation failure rate function in certain detection time points is determined. The lifetime reliability function of the product under the predetermined degradation failure threshold is obtained. An example is provided to demonstrate the validity of the proposed method.