Investigating the applicability of hybrid digital models for manufacturing systems under varying model input conditions

• Applicant: Professor Dr.-Ing. Jan C. Aurich
• Subject Area: Production Systems, Operations Management, Quality Management and Factory Planning
• Term: since 2019
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 418818038

The concept of the digital twin is seen as a core element of Industry 4.0, as it enables predictive, monitoring and diagnostic functions to be implemented using manufacturing data. Implementations of the digital twin at the factory, machine or process level enable a high degree of responsiveness in volatile and difficult-to-predict situations and thus contribute to an economic operation of a manufacturing system. Achieving these goals requires digital models of the respective real-world object of interest. In this context, hybrid digital models based on the integrated use of physics-based and data-driven models show a high potential, which could be shown in the first funding period of the proposed project. For the development and operation of such models in digital twins, the availability of sufficient model inputs in the form of data, information and knowledge is required. However, this prerequisite is not fully met in a large number of industrial manufacturing systems, which in turn makes it difficult for these companies to benefit from the potentials of digital twins. This especially puts manufacturing companies with limited implementation of Industry 4.0 in a disadvantageous position, which is often the case for small and medium-sized enterprises (SMEs).Therefore, in order to broaden the application scope of the digital twin to such scenarios, the applicability of hybrid models in environments with reduced availability of model inputs needs to be investigated and ways to overcome the resulting deficits between available and required model inputs need to be explored. Despite this need and the resulting potential, the systematic investigation of this research question has not been addressed in any approach so far.Therefore, based on exemplary use cases, the required and available model inputs for different hybrid models are described and the resulting gaps existing in this context are investigated. This forms the basis for the investigation of methods that enable the operation of the models while addressing the gaps. Methods for adjusting the models themselves, as well as the inputs for model operation, are explored. Finally, the methods and findings will be transferred into a conceptual framework and integrated with the procedure that was developed the previous funding period. Based on this, the final concept is validated with a further use case.

• DFG Programme: Research Grants
• International Connection: Brazil
• Partner Organisation: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior; (CAPES); Setor bancario Norte
• Cooperation Partner: Professor Dr. Fábio José Pinheiro Sousa