Geodesic
AI Large Models Bring Great Opportunities to Reusable Design of CAD Software
Computer Science and Information Systems, Tome 21 (2024) no. 4.

Voir la notice de l'article provenant de la source Computer Science and Information Systems website

With the rapid development of artificial intelligence (AI), large models have achieved significant breakthroughs in general-purpose domains. However, their application in computer-aided design (CAD) software is still in its early stages. Reusable design is crucial for improving efficiency and innovation in CAD systems. This paper reviews progress in rule-based reasoning (RBR) and case-based reasoning (CBR), two prevailing techniques for reusable design. RBR represents expert knowledge as rules but lacks self-learning capabilities. CBR draws on prior cases to solve new problems but relies heavily on surface empirical knowledge. Recent advances in large AI models provide new opportunities to enhance reusable design, thanks to superior language and reasoning abilities. However, adapting large models to effectively leverage CAD-specific design knowledge presents open challenges. To advance progress in this area, this paper analyzes the potential impacts of large models on improving knowledge acquisition, case retrieval, rule representation, and reasoning explain ability for hybrid CBR-RBR systems, and proposes a reusable design framework combing large language models, knowledge graphs, and databases to realize more intelligent and interpretable reuse. This review synthesizes key developments in RBR, CBR, and large AI models, highlighting promising directions for advancing reusable design in CAD software. The integration of reasoning techniques with large models, opening promising new directions for computer-aided engineering enhanced by artificial intelligence, as well as lays the foundation for more efficient, innovative, and sustainable engineering design.
Keywords: large model, CBR, RBR, reusable design 
@article{CSIS_2024_21_4_a17,
     author = {Yunlei Sun and Bingyi Yan and Zhaotong Shao},
     title = {AI {Large} {Models} {Bring} {Great} {Opportunities} to {Reusable} {Design} of {CAD} {Software}},
     journal = {Computer Science and Information Systems},
     publisher = {mathdoc},
     volume = {21},
     number = {4},
     year = {2024},
     url = {http://geodesic.mathdoc.fr/item/CSIS_2024_21_4_a17/}
}
TY  - JOUR
AU  - Yunlei Sun
AU  - Bingyi Yan
AU  - Zhaotong Shao
TI  - AI Large Models Bring Great Opportunities to Reusable Design of CAD Software
JO  - Computer Science and Information Systems
PY  - 2024
VL  - 21
IS  - 4
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/CSIS_2024_21_4_a17/
ID  - CSIS_2024_21_4_a17
ER  - 
%0 Journal Article
%A Yunlei Sun
%A Bingyi Yan
%A Zhaotong Shao
%T AI Large Models Bring Great Opportunities to Reusable Design of CAD Software
%J Computer Science and Information Systems
%D 2024
%V 21
%N 4
%I mathdoc
%U http://geodesic.mathdoc.fr/item/CSIS_2024_21_4_a17/
%F CSIS_2024_21_4_a17
Yunlei Sun; Bingyi Yan; Zhaotong Shao. AI Large Models Bring Great Opportunities to Reusable Design of CAD Software. Computer Science and Information Systems, Tome 21 (2024) no. 4. http://geodesic.mathdoc.fr/item/CSIS_2024_21_4_a17/