Article ID: | iaor2016353 |
Volume: | 31 |
Issue: | 8 |
Start Page Number: | 1553 |
End Page Number: | 1563 |
Publication Date: | Dec 2015 |
Journal: | Quality and Reliability Engineering International |
Authors: | Wang Fu-Kwun, Lu Yi-Chen |
Keywords: | maintenance, repair & replacement, simulation, combinatorial optimization, statistics: distributions |
In this study, a new model based on the mixture of bounded Burr XII failure intensity and bounded intensity process is proposed to describe the failure intensity of minimally repaired systems with approximate bathtub behavior. The estimates of the model parameters are easily obtained using the maximum likelihood estimation method. The confidence intervals for the model parameters are also provided. Other existing models, such as superposed power law process, log‐linear process–power law process, and bounded bathtub intensity process are used to compare with our proposed model. Through numerical examples, the results show that our proposed model performs well regarding the Akaike information criterion value and the mean of square errors.