Many software reliability growth models (SRGMs) have been proposed in the literature to evaluate the remaining faults and software reliability. The probability of getting failure-free software within a specified period and environment is known as software reliability and is recognized as one of the essential aspects. In this paper, we present a new SRGM of the fault detection process (FDP) and fault correction process (FCP) and study the dependency between the FDP and FCP as the amount of fault dependency, not time dependency. The FDP is modeled by considering a multi-release concept where the leftover faults from the previous release and newly added faults after enhancing the existing features in the software are considered. Further, the FCP model is proposed by introducing the change point concept in the fault dependency function. These models are validated on two actual medium-sized software system data sets. The results show that the proposed models fit the data set more accurately than the existing SRGMs. We have also discussed the optimal release time through a cost model where setup, testing, and debugging costs are introduced in both the testing and operational phases in the cost model.
Keywords: Fault correction process; Fault detection process; Mean value function.; Non-homogeneous Poisson process; Software reliability growth models.
© 2025. The Author(s).