Abstract
Morphological analysis is often used in the conceptual design process because it provides a structured and systematic approach to decompose design problems, improving the completeness and diversity of creative solutions. However, it still relies heavily on designers' professional knowledge and experience when decomposing design problems and proposing solutions, which is a challenge for designers. In this study, we propose an LLMs-augmented morphological analysis approach for conceptual design. We refine the design process as three main stages: decomposition, generation, and combination, offering targeted guidance and support to designers when applying morphological analysis. In addition, we introduced a feedback mechanism to develop more appropriate solutions to specific problems. Experimental results show that our approach improves the quality of innovative conceptual designs and reduces cognitive load. Furthermore, it augments the interactive experience between the designer and LLMs.
Keywords
large language models; conceptual design; morphological analysis; human-ai co-creation
DOI
https://doi.org/10.21606/drs.2024.605
Citation
Chen, L., Tsang, Y., Jing, Q., and Sun, L. (2024) A LLM-augmented Morphological Analysis Approach for Conceptual Design, in Gray, C., Ciliotta Chehade, E., Hekkert, P., Forlano, L., Ciuccarelli, P., Lloyd, P. (eds.), DRS2024: Boston, 23–28 June, Boston, USA. https://doi.org/10.21606/drs.2024.605
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Conference Track
Research Paper
Included in
A LLM-augmented Morphological Analysis Approach for Conceptual Design
Morphological analysis is often used in the conceptual design process because it provides a structured and systematic approach to decompose design problems, improving the completeness and diversity of creative solutions. However, it still relies heavily on designers' professional knowledge and experience when decomposing design problems and proposing solutions, which is a challenge for designers. In this study, we propose an LLMs-augmented morphological analysis approach for conceptual design. We refine the design process as three main stages: decomposition, generation, and combination, offering targeted guidance and support to designers when applying morphological analysis. In addition, we introduced a feedback mechanism to develop more appropriate solutions to specific problems. Experimental results show that our approach improves the quality of innovative conceptual designs and reduces cognitive load. Furthermore, it augments the interactive experience between the designer and LLMs.