Data Fabric: A Concept for Enhancing Data Management in Manufacturing

As manufacturing companies continue to adopt digital transformation, leveraging production data can significantly improve performance and drive innovation. Industrial Internet of Things technologies can help harness this data but they struggle with the complexity introduced by typical data sources in manufacturing. Different protocols, device semantics and inconsistent data standards lead to fragmented and inaccessible data. These obstacles prevent the seamless transfer of data from production resources to data-driven use cases, requiring manual effort to effectively use the data. To address this problem, this paper proposes a novel approach to data management based on the Data Fabric concept, which uses linked data principles to streamline data integration. By characterizing data pipelines from source to sink, consolidating metadata into a central knowledge graph, and employing an operational framework for configuring and managing these pipelines, our approach enhances data transparency and discovery. Utilizing this comprehensive strategy, we achieve the seamless provisioning of data for the implementation of data-driven use cases and provide a transformative solution for efficient and optimized data management in manufacturing.