Along with increasing digitisation, consistent data management is a decisive success factor, especially when several researchers in the scientific community work together. Implementing the FAIR data principles (Findable, Accessible, Interoperable, Reusable) ensures that existing research data is sensibly utilised. Thus, on the one hand redundancies, ambiguities and contradictions can be avoided. On the other hand, higher-level relationships can be recognised with the help of established or data-driven engineering methods.
Therefore, a Product-Lifecycle-Management (PLM) system has been introduced in a collaborative research centre with focus on mechanical joining. To ensure its applicability, the platform is adapted to the specific needs of the participating researchers from various domains. Therefore, the main purpose is the consistent long-term archiving and collaboration through the definition of explicit metadata. Moreover, workflows with predefined access and write permissions guarantee proper usage within the research project. This allows, for example, to access and reuse data from finite element simulation studies on the behaviour of the joints as training data for machine learning models predicting their results.
While the PLM system focuses on data, the proper documentation of existing knowledge also plays an important role. Therefore, a cross-domain ontology for mechanical joining processes has been developed that formally describes the knowledge from the different domains semantically. This explicit representation of relationships allows, for example, to derive qualitative statement on the joinability on the basis of the existing knowledge.
The future enhancement and, above all, linking of these two approaches holds great potential for increasing reusability and thus gaining new findings on mechanical joints. For example, the plausiblity of simulation results in the PLM sytem could be checked utilizing the relationships defined in the ontology with the help of SPARQL queries.