Optimal Release Policies of Delayed Discrete Reliability Growth Model With Imperfect Debugging

The present investigation was concerned with the delayed discrete S-shaped software reliability growth model with imperfect debugging. In this paper, we developed a discrete model based on non-homogeneous Poisson process by considering two types of fault removal processes. Optimal release policy was suggested to predict the optimal test run after which the software should be released. The sensitivity analysis was performed on the cost and reliability functions. We used the adaptive neuro fuzzy inference system approach to facilitate the computational results and compared it with the analytical results.