Automatic Publication of Data to Data- and Service Ecosystems from the Shopfloor

Abstract

In the manufacturing sector, vast volumes of data are generated at the field layer, offering valuable insights into the quality of produced parts, process efficiency, and various other applications. Traditionally, these data have been managed and analyzed through centralized infrastructures either in a cloud or on-premise. However, to tackle trust-related concerns from data providers and enable value creation through services by external companies, the concept of data spaces and ecosystems has emerged. Multiple drafts for standards for ecosystems exist; all offering different approaches for trustful cross-company collaboration. Notably, the Gaia-X initiative aims to establish a federated and sovereign data infrastructure. This paper introduces a novel approach for automatically publishing production data and associated services to multiple instances of such ecosystems directly from the shopfloor by deploying microservices on edge devices close to the machine. The developed architecture combines data acquisition of machine data through standards like OPC UA, creation of verifiable metadata description for the collected data sets, and publishing the metadata to different data and service ecosystems adhering to the Gaia-X standard, enabling trustful machine-to-machine connections. It is shown how services like data storage, ERP/MES, and asset management systems can be seamlessly deployed along the cloud edge continuum into this process. This approach fosters data integrity, sovereignty and enables coherent integration within the evolving cooperative manufacturing landscape.