A generalized framework for modelling multi up-gradations of a software with testing effort and change point

Abstract

The competitive essence of market and rapid change in human's requirement in technology is shrinking the age of softwares. Increasing demand of reliable software and stiff competition in the market has forced software companies for releasing their softwares in multiple versions. Software reliability is a major and leading factor for measuring the software quality, which is essential for consumer satisfaction. In last thirty years, various testing effort based Software Reliability Growth Models (SRGMs) have been proposed in the literature based upon fixed or monotonically growing Fault Detection Rate (FDR). In practice, as the testing grows, skill and efficiency of tester also grows. Also, due to the introduction of new testing strategies, new test cases, defect density, variation in resource allocation and running environment there comes a change in FDR. The point on time line the change in FDR is appears, termed as change point. In this paper we integrated the concept of change point with testing effort to propose a generalized modelling framework to model multi up-gradations of a software. Further, we have used Generalized Modified Weibull Distribution testing effort function for fault estimation. The prognosticative power and accuracy of the model has been work on a real set of failure data of tandem computers for four release.