A Scalable Edge Computing Solution for Real-Time Process Monitoring and Optimization in Industrial IoT

Digital transformation in manufacturing requires scalable solutions with real-time data processing capability, which is still constrained by the limitations of legacy solutions. Can this limitation be overcome by introducing edge devices in a modular and layered system architecture? The presented case study examines if and how edge devices can be seamlessly integrated into two different digital manufacturing laboratory systems, one academic and one industrial, using a predefined system architecture. This architecture, the Factory Real-Time Edge Innovation (FREI) architecture was developed by a Global Bearing Manufacturer (GBM), based on use cases defined for handling High Frequency Real-Time (HFRT) data and selected for their future potential for AI-based grinding process optimisation. In the academic laboratory environment, a force sensor was used to generate real-time HFRT data for a grinding process, which was then transferred to a cloud environment, by the edge-devices using industrial protocols. In the industrial lab, a PLC was used to simulate a grinding process instead of a physical grinding machine. The case study with the FREI architecture demonstrated the feasibility of using edge computing to connect a legacy grinding machine with external sensors, enabling real-time process monitoring and the seamless distribution of data processing tasks between edge and cloud environments, facilitated by standard communication protocols. The key takeaway from the case study was that while it is possible to seamlessly integrate an edge device into legacy system architectures, it can also require significant configuration effort.