Optimal software release policy with change-point
Shinji Inoue, Shigeru Yamada · 2008
Testing-time when a characteristic of a software failure-occurrence or fault-detection phenomenon is notably changed is ordinarily called change-point. The effect of the change-point on software reliability growth process influences on accuracy of the software reliability assessment based on SRGMs which are mostly developed under the assumption that the stochastic characteristic of the software failure-occurrence or the fault-detection phenomenon does not change throughout the testing. In this paper, we discuss a framework for software reliability growth modeling with change-point as one of the solutions to incorporate the effect of the change-point into software reliability assessment, and also discuss its application to an optimal software release problem with change-point. That is one of the interesting issues for project management of software development. Finally, we show numerical examples of our model and derived software release policy by using actual data.