Software release and testing stop time decision with change point

Abstract

Information technology has seen enormous growth in the last few decades. The backbone of this growth is the developments in the software industry who is working day and night to meet all the expectations. As we know a reliable software can ease our work but developing reliable software may take a longer time. At the same time, huge market competition has forced firms to deliver reliable products as quick as possible as per the user’s expectation. Due to this trade-off between reliability and market competition firms have started releasing software products with the assurance that they will repair their products if any failure is detected while operating. Keeping this phenomenon in mind various researchers proposed a framework based on a different set of assumptions. During the software development at some point of time the rate of detection/removal changes due to various reasons such as a change in testing efficiency, team composition, the addition of extra testing team, etc. This change in the fault removal has an impact on the release and testing stop time of software. In this paper, we propose a software cost model to determine optimal software release and testing stop time considering the change in detection/removal rate before the release of the software. The proposed model is validated on real-life data set.