AI-Driven Predictive Analytics for Enhancing Software Quality: A Review of Performance, Functional and Automation Testing
AI-Driven Predictive Analytics for Enhancing Software Quality: A Review of Performance, Functional and Automation Testing
Authors:
Snehal P. Vatturkar1, Dr. Brijendra Gupta2
1,2 Department of Information Technology, Siddhant College of Engineering, Savitribai Phule Pune University, Pune, Maharashtra, India
Abstract - This review investigates the role of Artificial Intelligence (AI) in enabling predictive analytics for performance testing, functional testing and automation testing within Software Quality Assurance (SQA). A systematic literature review was conducted following the PRISMA guidelines which covering peer-reviewed studies published between 2020 and 2025. Major scientific databases, including IEEE Xplore, ACM Digital Library, SpringerLink, Elsevier ScienceDirect and Scopus, were examined. An initial set of 1,120 records was identified from which more than 70 high-quality studies were selected after rigorous screening and eligibility assessment.
The review highlights significant advances in AI-driven test automation, predictive defect modelling, anomaly detection, performance prediction and regression suite optimization. Emerging techniques such as Large Language Models (LLMs), Graph Neural Networks (GNNs), Reinforcement Learning (RL), Explainable Artificial Intelligence (XAI) and federated learning demonstrate strong potential to transform traditional QA practices by enabling adaptive, scalable and data-driven testing processes. However, the analysis also reveals persistent challenges, including data imbalance, limited model generalizability, lack of standardized datasets, interpretability concerns, integration complexity within CI/CD pipelines and workforce skill gaps.
By synthesizing research trends across multiple testing domains, this review provides a comparative assessment of traditional and emerging AI techniques and identifies critical research gaps. Furthermore, it proposes a future roadmap toward scalable, interpretable and industry-ready AI-powered QA frameworks. Overall, the findings confirm that AI-driven predictive analytics is a rapidly evolving component of Quality Engineering, offering substantial improvements in software reliability, testing efficiency and alignment with modern business and development demands.
KeyWords: Artificial intelligence; Machine learning; Predictive analytics; Software quality assurance; Performance testing; Functional testing; Automation testing; Defect prediction