A Data Management Concept for Learning Factories to Support Scenario-Based Learning of Advanced Manufacturing Data Analytics for SMEs

Increasing digitalization has led to vast amounts of data being generated in small- and medium-sized manufacturing companies. Despite the potential for gaining deeper insights in operational performance and improving efficiency, many small- and medium-sized enterprises (SMEs) struggle to capitalize on data-driven insights, largely due to poor data analytics (DA) skills and scarcely documented use-cases. Despite initial efforts, i.a. in a learning factory (LF) context, the transition to advanced analytics, particularly machine learning techniques, is often hampered by a lack of hands-on learning offerings. Exploring the role that LFs can play in bridging this gap, this study illustrates potential benefits of merging process and machine data from LFs with different production settings. After selecting SME-specific use-cases through literature research and interviews with SMEs in the region, data management (DM) requirements to support teaching activities of advanced DA for SMEs in a scenario-based learning setting were assessed and relevant data from the LFs were identified. On this basis, a DM concept for our LFs to support teaching with data from different production systems was developed. We could show that our DM approach covers a wide range of SME-relevant DA tasks. Furthermore, we highlighted the contribution of collecting data from more production settings, considering also the prototyping phase of the product lifecycle and different manufacturing areas.