{"language":[{"iso":"eng"}],"publisher":"Faculty of Technical Science, Department of Industrial Engineering and Management ","publication_status":"published","date_created":"2024-11-18T10:20:34Z","type":"conference","user_id":"44204","year":"2024","author":[{"id":"44204","first_name":"Manuel","last_name":"Ott","full_name":"Ott, Manuel"},{"id":"38611","last_name":"Meihöfener","full_name":"Meihöfener, Niclas","first_name":"Niclas"},{"id":"95903","first_name":"Iryna","full_name":"Mozgova, Iryna","last_name":"Mozgova"}],"status":"public","department":[{"_id":"741"},{"_id":"144"}],"citation":{"chicago":"Ott, Manuel, Niclas Meihöfener, and Iryna Mozgova. “An Approach to Use Generic Data Sets for Neural Networks in Product Designs through Geometric Abstraction via Primitives.” In <i>Proceedings of the 11. Conference on Mass Customization and Personalization (MCP)</i>, edited by Zoran  Anisic and Cipriano Forza. Novi Sad, Serbia: Faculty of Technical Science, Department of Industrial Engineering and Management , 2024.","apa":"Ott, M., Meihöfener, N., &#38; Mozgova, I. (2024). An approach to use generic data sets for neural networks in product designs through geometric abstraction via primitives. In Z. Anisic &#38; C. Forza (Eds.), <i>Proceedings of the 11. Conference on Mass Customization and Personalization (MCP)</i>. Faculty of Technical Science, Department of Industrial Engineering and Management .","mla":"Ott, Manuel, et al. “An Approach to Use Generic Data Sets for Neural Networks in Product Designs through Geometric Abstraction via Primitives.” <i>Proceedings of the 11. Conference on Mass Customization and Personalization (MCP)</i>, edited by Zoran  Anisic and Cipriano Forza, Faculty of Technical Science, Department of Industrial Engineering and Management , 2024.","ieee":"M. Ott, N. Meihöfener, and I. Mozgova, “An approach to use generic data sets for neural networks in product designs through geometric abstraction via primitives,” in <i>Proceedings of the 11. Conference on Mass Customization and Personalization (MCP)</i>, Novi Sad, Serbia, 2024.","short":"M. Ott, N. Meihöfener, I. Mozgova, in: Z. Anisic, C. Forza (Eds.), Proceedings of the 11. Conference on Mass Customization and Personalization (MCP), Faculty of Technical Science, Department of Industrial Engineering and Management , Novi Sad, Serbia, 2024.","bibtex":"@inproceedings{Ott_Meihöfener_Mozgova_2024, place={Novi Sad, Serbia}, title={An approach to use generic data sets for neural networks in product designs through geometric abstraction via primitives}, booktitle={Proceedings of the 11. Conference on Mass Customization and Personalization (MCP)}, publisher={Faculty of Technical Science, Department of Industrial Engineering and Management }, author={Ott, Manuel and Meihöfener, Niclas and Mozgova, Iryna}, editor={Anisic, Zoran  and Forza, Cipriano}, year={2024} }","ama":"Ott M, Meihöfener N, Mozgova I. An approach to use generic data sets for neural networks in product designs through geometric abstraction via primitives. In: Anisic Z, Forza C, eds. <i>Proceedings of the 11. Conference on Mass Customization and Personalization (MCP)</i>. Faculty of Technical Science, Department of Industrial Engineering and Management ; 2024."},"title":"An approach to use generic data sets for neural networks in product designs through geometric abstraction via primitives","publication":"Proceedings of the 11. Conference on Mass Customization and Personalization (MCP)","conference":{"end_date":"2024-09-27","location":"Novi Sad, Serbia","name":"11. Conference on Mass Customization and Personalization (MCP)","start_date":"2024-09-24"},"editor":[{"first_name":"Zoran ","full_name":"Anisic, Zoran ","last_name":"Anisic"},{"full_name":"Forza, Cipriano","last_name":"Forza","first_name":"Cipriano"}],"place":"Novi Sad, Serbia","_id":"57182","date_updated":"2024-11-28T06:45:13Z","publication_identifier":{"isbn":["978-86-6022-686-2"]},"abstract":[{"lang":"eng","text":"Generative design suggestions and topology optimizations can help to reduce iterative process loops between calculation and design departments during product development processes. However, precise topology optimizations are computationally intensive, while generative designs benefit from swift suggestions to address design problems efficiently. Using artificial neural networks (ANN) can address this contrast of pre-defined aims by predicting topology-optimized designs, thereby combining both advantageous features. However, a challenge in Mass Customization is, that ANN are usually trained on specific geometries, making transfer to other applications impractical or requiring the creation of new datasets, which is economically unfeasible. Authors have already demonstrated a solution in other publications: dividing a geometry into geometric primitives like cuboids to perform abstraction. An ANN can then be trained to recognize optimized cuboids, which can be assembled back into a complete geometry, comparable to the finite element methods, which divide geometries of parts in finite elements enable mechanical property calculation. This publication aims to illustrate the steps of the approach in which the complete geometry of a part is segmented into these primitives, and the benefits obtained. Various methods will be explored, including automated workflows on modern low-code platforms, to enable generalized use."}]}