An NHPP Software Reliability Model Considering a Monotonic Decreasing Fault Detection Rate Function

Purpose: This study aims to improve software reliability growth models (SRGMs) by introducing a decreasing fault detection rate that reflects the temporal characteristics of real testing environments.<br/>Methods: A new non-homogeneous Poisson process (NHPP) SRGM is proposed, in which the fault detection rate is defined as a time-dependent decreasing function. Model parameters were estimated using the least squares method based on an actual failure dataset, and the proposed SRGM was compared with 10 traditional SRGMs across 9 evaluation criteria.<br/>Results: The proposed SRGM demonstrated the best model fit and predictive performance on both datasets.<br/>Conclusion: The proposed NHPP SRGM effectively captures the decreasing fault detection efficiency over time and provides more accurate reliability predictions than conventional SRGMs.