Towards a topological reasoning service for IFC-based building information models in a Semantic Web context

One of the classic problems identified in the interdisciplinary use of Building Information Models (BIM) is the different representation requirements regarding topology (Eastman 1999). Although this problem has been addressed in several modeling efforts (Augenbro 1995) the most widely spread BIM to date (IFC) does not bridge one of the essential gaps, namely that between geometry-centered (used by most generic CA(A)D applications) versus topological space-centered models (used by energy performance tools). Based on a previously developed Description Logic representation of the IFC model notated in OWL (Beetz, de Vries, van Leeuwen 2005), we describe the ongoing development of an online reasoning service to demonstrate the practical use of the Semantic Web tool chain in the AEC domain context. This reasoning service is part of a demonstration scenario in which energy performance estimates based on the ESP-r package are integrated into the architectural design process mediated by a multi agent system. In this paper we demonstrate how to infer spatial relations of a geometry-centered building model instance exported from standard packages. The approach we propose can be used in various scenarios where translation between different representations and mappings of other information are necessary. While these transformations and mappings are well understood and have been proven feasible in other research and developments (Bazjanac 2004, Treeck 2004), the uniqueness of our approach lies in the computational modularization of some aspects of this process and its integration into emerging technologies of the Semantic Web. Chaining these kind of semantically enhanced reasoning service modules together, future domain application developers might be relieved of implementing some of the cumbersome aspects when integrating their tools into the design and decision making process.