Search for a command to run...
Industry Foundation Classes (IFC) is a complete, wide and complex open standard data model to represent Building Information Models. Big efforts are being made by the standardization organization buildingSMART, to develop and maintain this standard in collaboration with researchers, companies and institutions. However, when trying to use IFC models from practice for automatic analysis, some issues emerge, as a consequence of a misalignment between what is prescribed by, or available in, the standard with the data sets that are produced in practice. In this study, a sample of models produced by practitioners for aims different from their explicit use within automatic processing tools is inspected and analyzed. The aim is to find common patterns in data set from practice and their possible discrepancies with the standard, in order to find ways to address such discrepancies in a next step. In particular, it is noticeable that the overall quality of the models requires specific additional care by the modellers before relying on them for automatic analysis, and a high level of variability is present concerning the storage of some relevant information (such as georeferencing).