Economic analysis of a stopping-rule in branch coverage testing

Abstract

This paper argues that software testing can be less thorough yet more cost-efficient if applied in a well-managed, empirical manner across the entire Software Development Life Cycle (SDLC). This is done by showing the cost-benefit analyses among other criteria. To ensure success, testing must be planned and executed within an Earned Value Management (EVM) paradigm as the experiment is conducted on a statistical-process controlled mindset. The Stopping Rule (MESAT) is applied to an actual embedded-chip software development cycle to show potential gains compared to archaic testing methods or none that were used. The result is that a considerable percentage of the particular testing effort could have been saved under usual circumstances, had the testing been planned and executed under EVM with the MESAT algorithm.