Optimal Release Policy for Software with Epistemic Uncertainty
Zhe Liu, Shihai Wang, Bin Liu, Rui Kang · Journal of Uncertain Systems · 2024
In this paper, finding the optimal release time of software is one of the challenging problems for software companies. Various optimal release policies have been proposed under the framework of probability theory, which is an axiomatic mathematical system dealing with aleatory uncertainties rather than epistemic uncertainties. Nevertheless, software faults embody lot of epistemic uncertainties that are more suitable to be modeled by uncertainty theory. Given this, a series of software belief reliability growth models under the framework of uncertainty theory were proposed, whose superiorities have been shown by several real data analyses. Up to now, the optimal software release policy based on software belief reliability growth models has not been studied comprehensively. As a consequence, in this paper, we discuss optimal release policy for software based on software belief reliability growth models. Meanwhile, a sensitivity analysis is conducted to investigate the effects of various model parameters on the optimal release time.