Artificial intelligence for quality assurance in friction stir welding – a review on opportunities and challenges
Jagadesh Kumar Jatavallabhula, Flavia Masubelele, Steadyman Chikumba, Vasudeva Rao Veeredhi
Abstract
Abstract Friction Stir Welding (FSW) arose as a game changing joining technology for high-strength materials, predominantly in aerospace, automotive, and marine applications. However, ensuring weld quality and process optimization remains a critical challenge owing to the complex interplay of parameters and the occurrence of defects. Latest advancements in Artificial Intelligence (AI) have shown tremendous potential in addressing these challenges, enabling predictive modelling, real-time monitoring, and adaptive control in FSW. The present work critically reviews the integration of AI techniques in FSW quality assurance, focusing on their opportunities and challenges. AI based methods for optimization of parameters, defect detection, and real-time quality prediction are also surveyed. Case studies showcasing the application of AI in industrial sectors, including aerospace and automotive, proving improvements in weld quality, productivity, and sustainability are also reviewed. Despite advancements in AI-driven FSW, challenges such as data scarcity, computational complexity, and the need for standardized frameworks hinder its widespread adoption. Future research should focus on developing hybrid AI models, reinforcement learning (RL) strategies, and Industry 4.0 integration to enhance process adaptability and reliability. This work highlights the transformative potential of AI in revolutionizing FSW processes, presenting a foundation for sustainable and intelligent manufacturing systems.