Exploratory Analysis of Generative AI in Automated CAD Design of Mechanical Parts

DOI

10.2991/978-94-6239-750-7_76How to use a DOI?

Keywords

Generative AI; Computer-Aided Design (CAD); OpenSCAD; Large Language Models (LLMs); 3D Modeling

Abstract

Modern manufacturing basically relies on CAD, yet building precise 3D models is still manual grind process which usually demands a lot of expertise. While Generative AI is beginning to address this through Text-to-3D, current models often prove insufficient for practical engineering. They tend to produce static meshes (like PLY or OBJ files) that look fine visually but lack detail and parametric history required for iterative design. This paper introduces CADgen an automated framework that leverages Large Language Models (LLMs) specifically speaking Google Gemini 2.5 Flash to generate editable, parametric OpenSCAD scripts from natural language prompts. We evaluate the system at range of complexity levels starting with simple structural primitives and moving up to mathematically complex mechanical parts. The results were mixed. While the model pretty much achieves a near 100% success rate for basic structural elements, it started to struggle with complex assemblies. That said, we identified that keeping a human-in-feedback-loop architecture significantly mitigated errors. It suggests that while LLMs may not yet be ready for fully autonomous operation, they demonstrate clear viability as a collaborative framework for engineering design. This study highlights the transformative role of LLMs in design workflows.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Shubham Yadav
AU  - Sumit Kumar
AU  - A. V. Muley
PY  - 2026
DA  - 2026/08/31
TI  - Exploratory Analysis of Generative AI in Automated CAD Design of Mechanical Parts
BT  - Proceedings of the International Conference on Advanced Design, Manufacturing, and Sustainable Energy Systems (ICADMSES 2026)
PB  - Atlantis Press
SP  - 1061
EP  - 1070
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6239-750-7_76
DO  - 10.2991/978-94-6239-750-7_76
ID  - Yadav2026
ER  -