A General Discrete Software Reliability Growth Model
P. K. Kapur, Said Younes, Sanjay Agarwala · International Journal of Modelling and Simulation · 1998
The authors develop a general discrete software reliability growth model based on nonhomogenous Poisson process. The software faults are categorized according to severity; corresponding effort is needed to detect and remove them. In other words, there may be delay in their detection and subsequent removal. The authors show that the model is flexible enough and fits well with a variety of actual data sets obtained from different software projects. Predictivity of the model is checked in terms of Akaike information criterion and sum of squared errors. The proposed model is also checked against different components of the model, including other existing ones, thus highlighting its applicability.