International Journal of High-Rise Buildings (국제초고층학회논문집)

Council on Tall Building and Urban Habitat Korea (한국초고층도시건축학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Development of an Artificial Intelligence-Based Architectural Agent for Schematic Design Concept Using GPT-4 Language Models and Case-Based Reasoning

  • Myung Sik Lee (Division of Architectural Engineering, Dongguk University) ;
  • Hyeok Joo Kwon (Division of Architectural Engineering, Dongguk University)
  • Published : 2025.12.25
Download PDF
⟨ Previous Next ⟩

Abstract

This study introduces the "Design Competition Concept Proposal AI Agent," designed to reduce cognitive load and enhance creativity in schematic design. In public competitions, architects often face information overload when interpreting briefs with legal, technical, and programmatic constraints. To address this, the system integrates GPT-4-based Large Language Models (LLMs) with Case-Based Reasoning (CBR) through a retrieve-reuse-revise framework. The process involves structuring brief information, extracting keywords across five categories-sense of place, program, functionality, sustainability, and publicness-and generating three differentiated concepts guided by spatial, environmental/technical, and socio-cultural strategies. Applied to five public projects, the system produced consistent keywords and diverse concepts, with expert reviews confirming reduced processing time and improved conceptual diversity. These findings highlight the potential of LLM-CBR integration to automate conceptual reasoning and support structured human-AI collaboration in early design.

Keywords

Acknowledgement

This work is supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Climate, Energy and Environment, Republic of Korea (RS-2024-00441420).

References

  1. Du, C., Esser, D., Nousias, S., & Borrmann, A. (2024). Text2BIM: Generating building models using a large language model-based multi-agent framework. arXiv. https://arxiv.org/abs/2408.08054
  2. Gero, J. S., & Roozenburg, N. F. M. (1997). Frameworks for creativity in design: A function behaviour structure model. Design Studies, 18(4), 425-437. https://doi.org/10.1016/S0142-694X(97)00018-2
  3. Jang, S., & Lee, G. (2024). Interactive design by integrating a large pre-trained language model and building information modeling. In Y. Turkan et al. (Eds.), Computing in Civil Engineering 2023 (pp. 291-299). ASCE. https://doi.org/10.1061/9780784485231.035
  4. Kolodner, J. L. (1993). Case-Based Reasoning. Morgan Kaufmann.
  5. Onatayo, D., et al. (2024). Generative AI applications in architecture, engineering, and construction: Trends, challenges, and future prospects. Architecture, 4(4), 46. https://doi.org/10.3390/architecture4040046
  6. Oxman, R. (2006). Theory and design in the first digital age. Design Studies, 27(3), 229265. https://doi.org/10.1016/j.destud.2005.11.003
  7. Paas, F., Renkl, A., & Sweller, J. (2003). Cognitive load theory and instructional design: Recent developments. Educational Psychologist, 38(1), 1-4. https://doi.org/10.1207/S15326985EP3801_1
  8. Reimers, N., & Gurevych, I. (2019). Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In Proceedings of EMNLP 2019, 3973-3983. https://aclanthology.org/D19-1410/
  9. Rittel, H. W. J., & Webber, M. M. (1973). Dilemmas in a general theory of planning. Policy Sciences, 4(2), 155-169. https://doi.org/10.1007/BF01405730
  10. Simon, H. A. (1973). The structure of ill-structured problems. Artificial Intelligence, 4(3-4), 181-201. https://doi.org/10.1016/0004-3702(73)90011-8
  11. Sweller, J., Ayres, P., & Kalyuga, S. (2011). Cognitive Load Theory. Springer.
  12. Yin, M., et al. (2023). An ontology-aided, natural languagebased approach for multi-constraint BIM model querying. Journal of Building Engineering, 76, 107066. https://doi.org/10.1016/j.jobe.2023.107066