Automatic Generation of Level of Detail for National Digital Twin Building Using Industry Foundation Classes

In this study, we propose a methodology for the automatic generation of building models from level of detail (LOD) 0 to LOD 3 that comply with KS X 6808-1, the national digital twin (NDT) building data model standard of South Korea, using industry foundation classes (IFC) data.This methodology enhances the utilization of 3D data derived directly from sensor-based spatial information acquisition technologies such as light detection and ranging (LiDAR) and photogrammetry.Previous studies have been limited to a specific LOD or remain at city geography markup language (CityGML) 2.0-based conversion.In contrast, we establish an integrated conversion framework optimized for the KS X 6808-1 standard based on CityGML 3.0 and we develop the IFC2NDTBuilding generator to evaluate the proposed methodology.Through the visualization verification of the converted data, we confirm that the structural consistency of the original model is maintained, and semantic preservation validation demonstrates a high preservation rate exceeding 98% in the experimental data.The results provide a practical method for efficiently constructing a national digital twin using IFC data and ensuring the consistency of 2D/3D geospatial information, thereby maximizing the utility of data obtained from advanced geospatial sensors.This methodology is expected to serve as a new approach to building information modeling (BIM)-geographic information system (GIS) integration.