Modeling software reliability growth under testing coverage, fault removal efficiency, and uncertain environments

Modern software powers critical systems in healthcare, transportation, finance, and defense. Ensuring software reliability is therefore a fundamental concern during development and testing. Reliability growth models have long been used to track fault detection trends and support release decisions, but many existing approaches assume simplified conditions. This paper introduces a software reliability growth model (SRGM) that addresses these limitations by integrating three essential factors: testing coverage, fault removal efficiency, and uncertainty in the operating environment. The model is formulated within a non-homogeneous Poisson process (NHPP) framework and employs a flexible coverage function together with an exponential distribution to capture environmental variability. Model parameters are estimated using the least squares method, and validation is performed with two real-world datasets. This study also examines the release policy by framing the release time decision as a multi-objective optimization problem. Using NSGA-II, the approach identifies release times that balance testing cost and reliability.