|
Title |
Prompt Engineering for LLM with Educational Theories for MEP Routing Tasks
|
|
Authors |
최원준(Wonjun Choi) ; 이상현(Sanghyun Lee) ; 김치경(Cheekyeong Kim) ; 허석재(Seokjae Heo) |
|
DOI |
https://doi.org/10.4334/JKCI.2025.37.6.687 |
|
Keywords |
대형 언어 모델; 프롬프트 엔지니어링; MEP; 경로 설계; 자동화 large language model; prompt engineering; MEP; routing design; automation |
|
Abstract |
This study examines the integration of educational theories into prompt engineering to enhance the performance of large language models (LLMs) in Mechanical, Electrical, and Plumbing (MEP) schematic design tasks within the construction industry. Drawing on parallels between human cognitive processes and LLM behavior, the research hypothesizes that prompts informed by educational principles can significantly improve the accuracy and quality of AI-generated outputs. Simplified representations of complex construction diagrams are utilized to demonstrate the applicability of LLMs in MEP design scenarios. The findings indicate that embedding educational strategies into prompt engineering enhances the relevance and precision of AI-generated designs, representing an initial step toward more effective AI-assisted design workflows in construction. This paper underscores the advantages of this approach and outlines potential avenues for future research to further refine and expand its application in the construction industry.
|