Automated generation of building energy model geometry from BIM-derived space and opening volumes

Accurate geometry is critical in Building Energy Models (BEM), particularly the second-level space-boundary (2LSB) surfaces that define thermal zone boundaries and support energy-transfer calculations. Existing 2LSB generation methods require detailed design input and are vulnerable to modeling errors and incomplete geometry, creating a persistent gap in automating the production of simulation-ready BEM geometry from Building Information Models (BIM). This paper addresses this gap by presenting an automated method for generating 2LSB surfaces directly from the space- and opening-volume data available in IFC files. The method requires only a user-defined construction-thickness threshold and intersection-free space volumes, while relying on the input space and opening data to be pre-processed and free of geometric inconsistencies. Its design enables scalable and robust handling of complex building geometries once properly prepared. Two case studies, a university building and an office building, demonstrate the approach’s scalability and effectiveness, including its accurate treatment of opening volumes. • Robust and efficient BEM geometry generation from minimal BIM inputs. • Requires only intersection-free space and opening volumes for reliable input. • Scales effectively to large buildings using federated BIM datasets. • Accurately processes complex and irregular space geometries.