Abstract

The Digital Twin idea in manufacturing engineering, the virtual representation of the product, promises smooth interaction between the variety of domains involved in the product development process as well as a clear traceability of decisions. To continue the Digital Twin throughout a product’s life-cycle, data standards are needed for the exchange of data and interoperability between different software landscapes of manufacturers and suppliers. While existing standards in CAD and CAE such as STEP AP242 and VMAP allow for smooth transfer of data within the two domains, simulation remains decoupled from the CAD-centred development process. Further development of the standards should therefore be based on the current industrial situation in the industry sectors and working positions concerned with the interaction of CAD and CAE. Thus, exploratory, semi-structured expert interviews were conducted in an industry survey. The results were used to update the perspective on the industrial situations and workflows. Although some attempts for a connection exist, they are presently not well established in industry. The main obstacles for cross-domain standardization are the differences on several layers: 1. Data discretization 2. Demands for processing and storage 3. Metadata 4. Management and workflow structures 5. Software feasibility and usage Additional issues faced by standardization, engineering, and management vary across and within industry groups. The result is a unidirectional data transfer from CAD to CAE, with only non-machine-interpretable 3rd level results being sent back to CAD. Existing commercial solutions are customized, expensive, and not used to their full potential. Therefore, workaround solutions, self-written codes, and Excel-spreadsheets prevail in places, where appropriate standardization might bring the two domains closer together, and slim down as well as speed up industrial workflows. These results imply, that a connection of the domains may be better approached by standardizing and integrating the workflow structures, rather than the model representations.