Change Point Reliability Modelling for Open Source Software with Masked Data Using Expectation Maximization Algorithm
Jianfeng Yang, Xibin Wang, Yujia Huo, Jing Cai · 2020
Masked data is a common missing failure data in reliability engineering. In this paper, multiple change points (CPs) software reliability growth model (SRGM) based on nonhomogeneous Poisson process (NHPP) is proposed using masked data. The C-Chart technology is used to estimate the position of the change point during the software failure process. Moreover, the maximum likelihood estimation (MLE) process of the model parameters is derived in detail, and Expectation Maximization (EM) algorithm is used to solve the likelihood function complicated problem. Finally, using the Tomcat 5 software failure data to conduct a comparative analysis of model performance, the results show that the proposed reliability model is useful and powerful.