Semantic Enrichment of a BIM Model Using Revit: Automatic Annotation of Doors in High-Rise Residential Building Models Using Machine Learning

Verifying that a building complies with fire safety regulations is a tedious, time-consuming and error-prone task that requires considerable human effort. To reduce this workload while improving accuracy, researchers have turned to Building Information Modeling (BIM) to automate the process.

Although BIM models often contain relevant data, they rarely meet all the requirements of (local) regulations. Almost all of the few automated code compliance checking tools available on the market require manual completion of data and annotation of the model. Automating this process helps to reduce repetitive work and human error.

In her article for Fire Technology, Soheila Bigdeli and her colleagues focus on evacuation route rules, which play an important role in building design. These compliance checks rely on accurate information about the location of exits, but this information is often missing in BIM models. In her research, she conducts a proof of concept for the use of simple and fast ML techniques such as KNN and XGBoost for the automatic annotation of low-level of detail (LOD) BIM models.

The "EvacRouteChecker" plugin was developed for Autodesk Revit software. The tool first uses the machine learning model to identify exit doors and then checks compliance with evacuation route regulations by highlighting non-compliant doors in the building model. The method can be applied to any object-oriented BIM tool capable of providing geometric data, including IFC files. It can also be extended to the automatic annotation of other elements in BIM models.

Read Semantic Enrichment of a BIM Model Using Revit: Automatic Annotation of Doors in High-Rise Residential Building Models Using Machine Learning in Fire Technology.