Software reliability modeling with different type of faults incorporating both imperfect debugging and change point

Abstract

In past four decades, many nonhomogeneous Poisson process (NHPP) based software reliability growth models (SRGMs) have been proposed to measure and assess the reliability growth of software. During the testing process, the faults which causes failure are detected and removed. One common assumption of many traditional SRGMs is that the fault removal rate is constant. In practical, the fault removal rate increases with time as learning and maturity of software engineer increases. Hence, time variant fault removal rate has been considered in this study. A complex software system may contain different category of faults. Some of faults can be easily detected and removed and some of faults required more effort to be detected and removed. Therefore, in this article, a NHPP based SRGM has been proposed which incorporates mainly two type of faults, major and minor. The concepts of imperfect debugging and change point have also been incorporated in the proposed SRGM. The parameters of the proposed SRGM is estimated using Statistical Package for Social Sciences (SPSS) software and validation of the proposed SRGM has been done using real life data set.