Smart additive manufacturing: An IoT-driven framework for predictive failure detection and sustainable operation

The convergence of Internet of Things (IoT) technologies with additive manufacturing is transforming 3D printing into an intelligent, adaptive process capable of self-monitoring and predictive control. This study presents a robust, edge-intelligent IoT architecture for real-time failure detection in fused deposition modeling systems. In the proposed architecture, a network of thermal cameras, vibration sensors, acoustic emission microphones, and current/voltage sensors streams data to an edge node that performs on-device deep learning inference and forwards condensed features to the cloud for further analytic. The system aims to detect, classify, and mitigate process anomalies before they cause print failures. Key components include a modular sensing layer, a low-latency edge-processing pipeline running on NVIDIA Jetson Orin, and a user-facing application layer for visualization and feedback. Experimental results demonstrate a failure detection of 92.4%, sub-50 ms response latency, and up to 78% reduction in material waste. Case studies from both industrial and open-source contexts further highlight practical deployments and sustainability benefits. These findings confirm that IoT-driven monitoring and predictive analytics significantly enhance the reliability, efficiency, and environmental performance of additive manufacturing.