Abstract:

The test results are always extrapolated from the test conditions to the usual usage conditions for reliability assessment. To improve the reliability assessment precision under varied environment conditions, a method of integrated reliability assessment based on proportional risk model is proposed. In this method, the proportional risk model is introduced to describe the relations between reliability and environment factors. Accordingto the data analysis or experience, the proportional risk model is reestablished. Then the log-likelihood function is treated as a generalized linear expression of Poisson variables, in which the indicator is described as a Poisson variable. With the generalized linear model, the maximum likelihood estimations of the model coefficients are obtained. Thus the influences of environment factors on the reliability of the product are measured quantificationally. The instance analysis shows that this integrated reliability assessment method improves the precision of reliability assessment using the varied environment test data synthetically and is straightforward for engineering application.