SmartEM aims to address the limitations of current engineering models by developing a reference architecture for engineering model spaces. The architecture will enable the reuse, exchange and integration of computational engineering models, reducing the need for costly design corrections and promoting early data and model exchanges. SmartEM will use AI-assisted methods to create surrogate models from heterogeneous data sources and allow their re-combination within a given engineering domain. The project will develop use-case model spaces to manage reusable and transferable engineering models for various domains and provide solutions for IP management to enable model exploitation in an increasingly digital engineering market.

Expected solutions and business benefits: Inspired open data-space concepts like Gaia-X, SmartEM will design a reference architecture for engineering model spaces. Model spaces are realized to organize reusable and transferable technical models for a specific application domain. SmartEM will develop AI-assisted methods to create surrogate models from heterogeneous data sources, and support their re-combination within a given domain. A surrogate model is a (black-box) engineering model that may replace measurement data or a (white-box) first-principle engineering model from which the surrogate model is automatically derived. The reference architecture provides rules for model labeling and governance concepts to increase the usability of model spaces. IP management will enable exploitation of models in an increasing model market. Model space implementations will be realized for as internal use-case solutions, as open community stores for e.g. wind power plants, or as commercial development platforms. A main business benefit of SmartEM solutions is that they allow to significantly reduce development times by recycling, recombining or re-training their first-principle, data-based, and hybrid models. Furthermore, the deeper technical cooperation with partner companies enabled by the SmartEM Model Store platforms can lead to completely new design and product approaches.
 

Project Partners

• Katholieke Universiteit Leuven, Belgium
• Siemens Industry Software NV, Belgium
• Canon Production Printing Netherlands B.V., The Netherlands
• Eindhoven University of Technology, The Netherlands
• Philips Consumer Lifestyle B.V., The Netherlands
• Reden BV, The Netherlands
• SemLab, The Netherlands
• Thermo Fisher Scientific, The Netherlands
• TNO, The Netherlands
• Acd Bilgi Islem ltd.sti., Türkiye
• Alpata Technology, Türkiye
• Inovasyon Muhendislik, Türkiye
• Fraunhofer SCAI, Germany

Project duration: 04/2024 until 03/2027