The increasing adoption of Industry 4.0 technologies has led to the need for efficient and scalable solutions for real-time quality monitoring in distributed manufacturing environments. Traditional cloud-centric systems often suffer from high latency, limited scalability, and network dependency, making them unsuitable for time-critical industrial applications. To address these challenges, this paper proposes a Multi-Agent Edge–Cloud System (MAECS) for distributed quality monitoring, integrating edge computing, cloud intelligence, and autonomous multi-agent coordination. In the proposed framework, edge nodes perform real-time defect detection using deep learning models, while cloud servers handle global analytics, model updates, and long-term optimization. A multi-agent architecture enables decentralized decision-making, dynamic task allocation, and efficient resource utilization across the system. The agents collaborate to optimize latency, accuracy, and energy consumption in heterogeneous environments. Experimental evaluation demonstrates that the proposed MAECS significantly outperforms conventional approaches. The system achieves a detection accuracy of 97.3%, compared to 91.2% in cloud-only systems and 93.5% in edge-only systems. Additionally, the proposed approach reduces processing latency to 35 ms, representing a substantial improvement over 250 ms in cloud-based systems and 80 ms in standalone edge solutions. The results confirm that integrating multi-agent coordination with edge–cloud computing enhances both performance and scalability. The proposed system provides a robust and efficient solution for real-time distributed quality monitoring and has strong potential for deployment in smart manufacturing and other industrial IoT applications.
Malik et al. (2026) studied this question.