Semantic enrichment of object associations across federated BIM semantic graphs in a common data environment

Explicit association semantics across federated BIM models facilitate the interoperability between design systems and open opportunities for intelligent applications. However, the automated inference and generation of such semantics remains a key challenge. Graph representations of BIM models have shown potential in supporting semantic enrichment and conveying formal building semantics across domains. In this paper, we present our implementation of a semantically-enriched Common Data Environment – the Graph-based Core-extension Data Framework – composed of a semantic graph layer and an object geometry extension layer, hosting the heterogeneous information from BIM models. A novel rule-based inferencing approach, the TIOC algorithm, was implemented on the federated building representations to infer and establish inter-domain topological and correspondence relationships, interconnecting disjoint domain specific graphs with meaningful formal semantics. This study is the first known attempt to implement a semantic enrichment algorithm for implicit object associations in the context of multidisciplinary models.