Methodological framework for evaluating the generative capabilities of AI assisted CAD parametric modeling for AM
Nicolae Răzvan MITITELU, Mihaela NICOLAU, Alexandru Ionuț IRIMIA, Cosmin-Gabriel GRĂDINARU
Abstract. This study proposes a theoretical methodology for the analysis of the performance of artificial intelligence systems in the automated generation of parametric CAD models, focusing on additive manufacturing. In addition, it provides a steps-based method that includes the formulation of metrics, testing, statistical analysis, error mapping, and formulation of recommendations. This method also provides a composite risk indicator that is able to provide the quantification of limitations, as well as support the formulation of decisions in industry. This method is scalable, portable, and can be applied for other applications such as topological optimization and generative design. Apart from its applicability, this research provides the first theoretical framework within which the standardization of artificial intelligence processes within CAD models can be achieved.
Keywords
Artificial Intelligence, 3D Design, Generative Design, AI CAD Design
Published online 1/20/2026, 7 pages
Copyright © 2026 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA
Citation: Nicolae Răzvan MITITELU, Mihaela NICOLAU, Alexandru Ionuț IRIMIA, Cosmin-Gabriel GRĂDINARU, Methodological framework for evaluating the generative capabilities of AI assisted CAD parametric modeling for AM, Materials Research Proceedings, Vol. 61, pp 77-83, 2026
DOI: https://doi.org/10.21741/9781644903995-10
The article was published as article 10 of the book Innovative Manufacturing Engineering and Energy
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
References
[1] I. Goodfellow, Y. Bengio & A. Courville, Deep Learning. MIT Press, 2016.
[2] A. Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems. O’Reilly Media, 2019.
[3] S. Russell & P. Norvig, Artificial Intelligence: A Modern Approach (4th Edition). Pearson, 2020.
[4] Y. Zhao, X. Gao & Z. Lin, A Review of Deep Learning Techniques for Additive Manufacturing. Journal of Manufacturing Processes, 77, 375–389. doi.org/10.1016/j.jmapro.2022.03.017, 2022.
[5] Y. Chen & Y. Li, Design automation for additive manufacturing using generative design and machine learning. Computers in Industry, 2021.
[6] Y. Zhang, L. Sun, C. Wang & Z. Liu, Modular Neural Network Architectures for Engineering Design Automation. Advanced Engineering Informatics, 2022.
[7] A. P. Schoellig, A. Tsiamis, M. Müller & A. Papachristodoulou, Human-in-the-loop AI for engineering design: A survey of recent advances. Annual Reviews in Control, doi.org/10.1016/j.arcontrol.2020.06.006, 2020.
[8] H. Bui, D. Phung, D. Do & B. Vo, Loss Function Engineering for High-Precision 3D Geometry Generation in Industrial Design. Computer-Aided Design, 136, 103030. doi.org/10.1016/j.cad.2021.103030, 2021.
[9] G. Cheng, S. Liu & Y. Wu, AI-Driven Topology Optimization: Advances and Challenges. Structural and Multidisciplinary Optimization, 67(4), 2023.
[10] Y. Wang, Y. Sun, Z. Liu, S. Sarma, M. Bronstein & J. Solomon, Dynamic Graph CNN for Learning on Point Clouds. ACM Transactions on Graphics, 35, 2019. https://doi.org/10.1145/3326362
[11] L. Mescheder, M. Oechsle, M. Niemeyer, S. Nowozin & A. Geiger, Occupancy Networks: Learning 3D Reconstruction in Function Space. CVPR 2019. https://doi.org/10.1109/CVPR.2019.00459
