Next-Generation SRGMs: A Unified Framework for Modeling Uncertainty, Testing Effort, and Intelligent Estimation in Complex Software Systems

Indarpal Singh, Sanjay Kumar, Arvind Kumar

Abstract: Software reliability has evolved into a critical measure of success for modern software-intensive systems, which now permeate safety-critical domains, blockchain ecosystems, and distributed environments. Software Reliability Growth Models (SRGMs), particularly those based on Non-Homogeneous Poisson Processes (NHPPs), have long been a foundation for quantifying the fault detection process over time. However, emerging complexities-including uncertain testing conditions, variable testing effort, imperfect debugging, and the advent of intelligent estimation techniques-require a comprehensive reconceptualization of SRGMs. This paper proposes a unified framework that integrates modern advancements in SRGMs, including the use of extended probability distributions (such as the Shanker and extended log-logistic models), dynamic testing effort modeled by Weibull functions, and intelligent prediction techniques encompassing neural networks, Bayesian inference, and fuzzy logic. Through a synthesis of theoretical models and empirical evidence, we demonstrate how these next-generation SRGMs outperform classical models across real-world datasets, particularly in blockchain-based implementations and testing environments with change-points. The unified framework presented not only strengthens model interpretability and estimation accuracy but also addresses the need for adaptive reliability prediction in agile and DevOps-centric workflows. This research ultimately contributes toward bridging the gap between theoretical modeling and practical reliability assessment in complex software systems.

Keywords: Software Reliability Growth Models (SRGMs), Unified Framework, Uncertainty Modeling, Testing Effort, Intelligent Estimation, Machine Learning in Software Reliability, Bayesian Estimation, Reliability Prediction, Complex Software Systems, Fault Detection and Removal, Reliability Engineering, Software Quality Assurance, Data-Driven Reliability Modeling and Artificial Intelligence in Software Testing, Predictive Analytics

How to Cite?: Indarpal Singh, Sanjay Kumar, Arvind Kumar, "Next-Generation SRGMs: A Unified Framework for Modeling Uncertainty, Testing Effort, and Intelligent Estimation in Complex Software Systems", Volume 14 Issue 10, October 2025, International Journal of Science and Research (IJSR), Pages: 1200-1212, https://www.ijsr.net/getabstract.php?paperid=SR251004124551, DOI: https://dx.doi.org/10.21275/SR251004124551