Linking ontology metrics with BIM modeling stages

Before starting a project, stakeholders of the Architecture, Engineering and Construction (AEC) industry aim to agree on common requirements, such as which information and geometry should be provided at any given stage of the engineering process. In particular, this consensus includes the geometric depth of the processed building information model and its enhancement with information artifacts. A suitable determination of a categorical standard can be achieved on the basis of Level of Geometry (LOG) and Level of Information (LOI). Since no international standard exists for these qualifiers, individual catalogs and checklists are currently used, which often do not allow an unambiguous classification. Given the subjective interpretability of existing methods, a more explicit and objective way of categorization would be beneficial. The purpose of this paper is to investigate whether this process can be supported by a mathematically rigorous methodology. If LOG and LOI could be determined by a formal computational process, the influence of subjectivity would be reduced to introduce a more objective categorization. With the proposed method, stakeholders can achieve a more reliable assignment of criteria and a common understanding. We examine whether the Resource Description Framework (RDF) representation of a building information model in Industry Formation Classes (IFC) format allows conclusions about categorization levels. Existing metrics for the analysis of ontologies are applied to RDF graphs. We then assess their correlation with LOG and LOI levels. Finally, we evaluate these mathematically objective qualifiers in terms of their relevance to the AEC domain. Our results show a promising framework for further exploration of a new class of BIM tools that facilitate collaboration and standardization.