Publications / 2021 Proceedings of the 38th ISARC, Dubai, UAE

BIM-based variant retrieval of building designs using case-based reasoning and pattern matching

Daniel Napps, Dennis Pawlowski and Markus König
Pages 435-442 (2021 Proceedings of the 38th ISARC, Dubai, UAE, ISBN 978-952-69524-1-3, ISSN 2413-5844)
Abstract:

Architects today still very often create traditional 2d plans in the early planning phase, which form the basis for discussions with investors, clients and engineers. The designs are often difficult to understand for non-experts and design variants can only be compared with great effort. The use of BIM models simplifies both communication and the actual creation through the use of parametric modeling approaches. Even though a building project always has very individual characteristics, it makes sense to take experience from other projects as a basis for the design. Therefore, it may be that suitable floor plans of similar buildings already exist, which are a good basis for variant studies in early design phases. A challenge is to be able to access the experience of other architects. Currently, this information is not available and is difficult to transfer from one project to the next. This paper outlines a solution for a BIM-based variant retrieval of a building design in early design phases using Case Based Reasoning (CBR) and Pattern Matching (PM). On the one hand, this offers the possibility to learn from old mistakes, and on the other hand, it enables the quick selection of suitable variants for a building from a diverse pool of variants.

Keywords: Building information Modeling (BIM); Case based reasoning (CBR); Early Design Phases; Industry Foundation Classes (IFC); Pattern Matching; Variant Retrieval; Similarity