In the "Neufert 4.0" project, artificial intelligence methods are being developed to derive knowledge (so-called design heuristics) from existing apartment floor plans, which helps to efficiently design residential buildings that are suitable for their use. The basis for this is a database with about 35,000 apartment floor plans. In addition to the geometry of the apartments, this database contains semantic information on building components, rooms and urban context as well as detailed information on use-related qualities (daylight, visibility, furnishability, spatial relationships). With this extensive database, correlations between floor plan geometry, urban context and use-related qualities are identified. For this purpose, different machine learning methods (regression models, Bayesian Networks, Generative Adversarial Networks) are used and tested for their suitability for the generation of design heuristics. Furthermore, visualization methods are developed to present the heuristics in an understandable way. In workshops with practicing architects these methods are tested for their suitability in concrete design situations.

Link to our research partner: Archilyse AG (https://www.archilyse.com/)