How Can Large Language Models Help Humans in Design And Manufacturing? Part 2: Synthesizing an End-to-End LLM-Enabled Design and Manufacturing Workflow
Abstract
Generative AI models provide foundational knowledge and sufficient reasoning to aid in individual aspects of a computational design and modeling workflow (see Part 1). In this work, we analyze the ability of state-of-the-art LLMs (in particular GPT-4) to reason about the entire end-to-end workflow, from conceptualizing to realization, for two target domains: static physical objects (here, furniture), and dynamical cyberphysical systems (here, quadcopters). Our investigation serves as an exemplar of how to use LLMs to build out such workflows, as well as a means of identifying which steps are most brittle and especially warrant future study. We conclude with an analysis of the capabilities and limitations of LLMs in the context of design and manufacturing as a whole, as well as a brief discussion of the grander opportunities and ethical risks put forth by such systems.
Acknowledgements
This material is based upon work supported in part by Defense Advanced Research Projects Agency (DARPA) Grant No. FA8750-20-C-0075, DARPA Fellowship Grant No. HR00112110007, and the National Science Foundation (NSF) under Grant No. 2141064. Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors(s) and do not necessarily reflect the views of the National Science Foundation.
Citation
@article{Makatura2024b_LLMsForCDAM,
author = {Makatura, Liane and Foshey, Michael and Wang, Bohan and H{"a}hnlein, Felix and Ma, Pingchuan and Deng, Bolei and Tjandrasuwita, Megan and Spielberg, Andrew and Owens, Crystal and Chen, Peter Yichen and Zhao, Allan and Zhu, Amy and Norton, Wil and Gu, Edward and Jacob, Joshua and Li, Yifei and Schulz, Adriana and Matusik, Wojciech},
journal = {Harvard Data Science Review},
number = {Special Issue 5},
year = {2024},
month = {dec 23},
note = {https://hdsr.mitpress.mit.edu/pub/hiii8fyn},
publisher = {The MIT Press},
title = {{How} {Can} {Large} {Language} {Models} {Help} {Humans} in {Design} {And} {Manufacturing}? {Part} 2: Synthesizing an {End}-to-{End} {LLM}-{Enabled} {Design} and {Manufacturing} {Workflow}}
}