Software reliability growth models incorporating change point with imperfect fault removal and error generation

The number of faults removed need not always be same as the number of failures observed in real software development environment. If number of failures observed is more than number of faults removed then we have the case of imperfect debugging. Due to complexity of software system and incomplete understanding of software, testing team may not be able to remove the fault perfectly on detection of failure and the original fault may remain leading to imperfect fault removal, or get replaced by another fault/error causing error generation. Attempts have been made to study the above cases separately. Most of the SRGMs are based upon constant or monotonically increasing fault detection rate (FDR). In practice, as testing grows, so does skill and efficiency of the testers. With introduction of new testing strategies and new test cases, there comes a change in FDR. The time point where the change in removal curve appears is termed as "change point". In this paper, we incorporate the concept of change point in software reliability growth in presence of imperfect fault removal and error generation. The models have been validated, evaluated and compared with other existing NHPP models by applying it on actual failure/fault removal data sets cited from real software development projects.