Unveiling the Evolution: Multi-Patch Multi-Release Software Reliability Growth Model with Testing Effort
Veenu Singh, Vijay S. Kumar, V. B. Singh, Arun Prakash Agrawal · Journal of Reliability and Statistical Studies · 2025
Reliance on software has increased expectations for software organizations to deliver high-quality software to meet the increasing demand from end-users. Continuous testing is imperative to ensure software quality, yet prolonged testing can lead to increased market opportunity costs. Consequently, organizations often opt to release software early and subsequently conduct testing during the operational phase, addressing existing bugs through patch deployment. These patches, small programs aimed at fixing, improving, or updating software, serve to rectify security vulnerabilities or bugs efficiently. For minor changes, patch releases prove more practical and cost-effective than launching entirely new software versions. The adoption of multi-release software endows developers with a competitive advantage, catering to the diverse needs of end-users. This paper introduces a testing effort-based software reliability growth model, evaluating the impact of multi-patching on multi-release software. The model operates under the assumption of continuous fault removal post-release, using different distribution functions to construct three framework variations. Parameter estimation employs the Statistical Package for Social Sciences, with a real dataset serving as a basis for a numerical example illustrating the model’s practical application. Additionally, a comparative analysis of model performance, based on different distribution functions, is conducted through multi-criteria decision-makingtechniques.