Ensuring zero-defect production in smart manufacturing demands fast, accurate, and intelligent inspection systems. However, conventional cloud-based defect detection approaches struggle with high latency, excessive bandwidth usage, and delayed decision-making, limiting their effectiveness in real-time industrial environments. To overcome these challenges, this paper proposes a novel Edge AI-driven framework for real-time defect detection, where optimized deep learning models are deployed directly on edge devices for instant analysis at the production line. The proposed system seamlessly integrates industrial vision sensors, edge computing units, and cloud platforms to achieve scalable, efficient, and intelligent quality control. Extensive experiments conducted on the MVTec AD dataset demonstrate that the proposed model achieves a high detection accuracy of 96%, while maintaining an ultra-low inference latency of 20 ms, significantly outperforming traditional cloud-based systems with latency exceeding 150 ms. Furthermore, the framework reduces bandwidth consumption by approximately 60%, enabling faster response times and efficient resource utilization. These results highlight the effectiveness of the proposed approach in delivering low-latency, high-accuracy, and scalable defect detection, making it a promising solution for next-generation Industry 4. 0 manufacturing systems. \\
Malik et al. (Thu,) studied this question.