Accelerating circular cities with semi-automatic building information modeling for existing buildings

The lack of high-granularity information on existing building stock impedes efficient material reuse and recycling, which is essential for promoting the circular economy in the building industry. Building Information Modeling (BIM) can fill this gap by offering detailed information about the material composition of buildings. However, creating BIM models for existing buildings remains costly, time-consuming, and labor-intensive. This study proposes a novel method for accelerating the creation of georeferenced BIM models by integrating heterogeneous mass data and domain knowledge about building construction. The method achieves 95% accuracy in estimating material intensities for external load-bearing timber frame walls and roof components. This approach provides valuable insights for stakeholders, facilitating the transition to circular cities by supporting material stock analysis at both macro and micro levels. By expediting the BIM modeling process, this method enhances the granularity of material stock analysis, supports efficient material reuse and recycling, and promotes urban sustainability. • The method enables semi-automatic georeferenced BIM models using heterogeneous data. • It helps AECO adopt circular economy principles through detailed component information. • Highlights specificity and context importance for material intensity calculations. • BIM offers high-granularity data on building material intensities. • BIM facilitates reuse, policymaking, and urban sustainability.