Abstract To improve the prediction accuracy of the lifetime of a certain type of missile connector, a prediction method synthesizing degradation data and lifetime data was proposed. By modeling the degradation data with Wiener process, modeling a small amount of lifetime data with inverse Gaussian distribution, establishing the reaction rate model with the generalized Eyring model, a comprehensive reliability model was established. Furthermore, the point estimates of the model parameters were obtained through the Markov Chain Monte Carlo simulation method, and the interval estimates were obtained through a bootstrap method. By comparing the prediction results, it was concluded that the approach synthesizing degradation data and lifetime data could improve the prediction accuracy.
Key words:
|