AI-Based Retrival to Encourage Reuse of CAD-Designs: A Methodological Study and Future Perspectives
DS 130: Proceedings of NordDesign 2024, Reykjavik, Iceland, 12th - 14th August 2024
Year: 2024
Editor: Malmqvist, J.; Candi, M.; Saemundsson, R. J.; Bystrom, F. and Isaksson, O.
Author: Gronvald, Jakob Meinertz; Norgaard, Morten; Heidari, Negar; Bakhtiarnia, Arian; Mortensen, Niels Henrik
Series: NordDESIGN
Institution: Technical University of Denmark; Aarhus University
Page(s): 277-283
DOI number: 10.35199/NORDDESIGN2024.30
ISBN: 978-1-912254-21-7
Abstract
Product development today faces challenges like customization demands and shorter life cycles. Companies adopt generational approaches to save time and costs but struggle with increased variation and opacity in product portfolios. This paper proposes using AI, specifically UV-Net, to address these issues by converting CAD solids into tensors for classification and retrieval tasks. Experimental results show promise, but challenges remain. Integrating cost considerations and further research can enhance AI-driven design assistance, aiding decision-making in product development.
Keywords: Artificial Intelligence (AI), Computer Aided Design (CAD), Information Retrieval, Geometric Feature Analysis, Reuse