A time-variant fault detection software reliability model

Abstract In this work, we propose a time-variant software reliability model (SRM)which considers the fault detection and highest number of faults in software. The genetic algorithm process is implemented for assessment SRM parameters. proposed works upon non-homogeneous Poisson (NHPP) incorporates dependent failure intensity un-removed error We had considered programmers proficiency, complexity, organization hierarchy, perfect debugging as determining factors SRM. dataset collected from 74 projects was experimented with to establish validate model's better fit. Data over period, which initiated start project continuously monitored until its completion. Several parameters are analyzed, collection 115 attributes given 11 different time frames terms product characteristics. A total 383 persons were involved design, where issue count 255. Jira also compared existing presented literature. It observed that like mean square (MSE), root (RMSE), r-squared (R 2 ). work carried out, ensuring time-varying detection, measured by considering response count, coding non-coding deliverables, bugs programmer's presenting model. Software shows improvement algorithms residual errors reduced, prediction accuracy high cumulative detection.