Automated Modeling Approach Based on the Spatial Object and its Property - Development of the Auto-Modeling of Residential Components Reflecting Users’ Physical Conditions -
This study proposes a high-LOD BIM automated design approach that effectively incorporates user requirements, particularly physical characteristics.The aim of the research is to systematize extensive design knowledge in order to minimize standardized design methods and repetitive tasks.Additionally, it aims to develop a practical tool capable of automatically reflecting design requirements based on user characteristics.To achieve this, the study integrates a pre-trained generative AI with an automated system, enabling automated design based on a user-object mapping table.The proposed system takes low-LOD BIM models and user spatial requirements as input and generates high-LOD BIM models automatically.The system is structured into three main modules that build and transfer design information, ensuring interoperability.To validate its effectiveness, the study conducted tests targeting elderly users with varying physical characteristics.Four scenario tests were performed in bathroom and kitchen spaces.The results demonstrated that BIM objects suitable for each user's aging characteristics were automatically selected and placed in the spatial models.This suggests that as design data for other user groups is developed, the system could be applied to a broader range of users.The significance of this research lies in proposing a practical tool that rapidly generates spaces tailored to individual user requirements.Furthermore, the study expands the applicability of BIM automation in user-centered design, contributing to improved efficiency in the detailed design phase of architectural processes.
Authors
- Jin-Kook Lee
- Su-Min Chae
Publication Details
- Journal
- Journal of the Korean housing association
- Published
- 2025-04-25
- DOI
- https://doi.org/10.6107/jkha.2025.36.2.087
- Primary Topic
- 3D Modeling in Geospatial Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00