ECC4P

The Fraunhofer project "Edge Cloud Continuum for Production" (ECC4P) uses data space technologies for the secure exchange of data between the edge and the cloud to optimize the manufacturing process of workpieces in production. Sensors monitor the manufacturing process on machine tools. By evaluating their data in a cloud, specialized ML models are generated that then optimize the control of the machines. This significantly reduces the reject rate and prevents machine downtime as much as possible. The parallel project "SoundTrack" uses the same scenario within the quality control of workpieces using newly developed sensors on a test bench for workpieces.

The challenge

Production processes generate large amounts of data, most of which remains unused. In order to use this data effectively in process optimization, it must be collected in parallel with the process at the edge and transferred to the cloud after individual release. There, machine learning (ML) models are trained with the data from the edge to derive action decisions that can be implemented directly by the operating personnel at the machine tool on site. It is of utmost importance that all data exchange is carried out confidently and securely and that control over the data always remains with the owner.

Our contribution

Within the framework of ECC4P, the individual systems and components were integrated into an »edge cloud continuum.« To do this, we use well-known data space technologies such as the open source »Eclipse Dataspace Components Connector« (EDC) and developed corresponding extensions to connect all data sources and sinks. The entire data flow between the edge and the cloud is controlled via customized web user interfaces in the data sharing, data transfers, training of ML models, and configuration of machine tools can be carried out by authorized personnel.

Results

The modularly developed technology package offers intelligent collection and administrative staff of data from production processes for further analysis by integrated AI pipelines. Data transfer is secure and reliable at all times and can be monitored and controlled via graphical user interfaces. The entire scenario remains flexible and can be adapted to existing systems or machines and expanded with new components such as additional sensors. The results are reduced scrap rates and optimized maintenance strategies that lead to more efficient and sustainable production processes.

Partners

• Fraunhofer- Institute for Machine Tools and Forming Technology IWU 
• Fraunhofer- Institute for Intelligent Analysis and Information Systems IAIS
• Fraunhofer- Institute for Integrated Circuits IIS
• Fraunhofer Institute for Applied and Integrated Security AISEC