IfcWoD, Semantically Adapting IFC Model Relations into OWL Properties

In the context of Building Information Modelling, ontologies have been identified as interesting in achieving information interoperability. Regarding the construction and facility management domains, several IFC (Industry Foundation Classes) based ontologies have been developed, such as IfcOWL. In the context of ontology modelling, the constraint of optimizing the size of IFC STEP-based files can be leveraged. In this paper, we propose an adaptation of the IFC model into OWL which leverages from all modelling constraints required by the object-oriented structure of IFC schema. Therefore, we do not only present a syntactic but also a semantic adaptation of the IFC model. Our model takes into consideration the meaning of entities, relationships, properties and attributes defined by the IFC standard. Our approach presents several advantages compared to other initiatives such as the optimization of query execution time. Every advantage is defended by means of practical examples and benchmarks.

💡 Research Summary

The paper addresses a long‑standing limitation in BIM (Building Information Modeling) interoperability: the way IFC (Industry Foundation Classes) relationships are represented in OWL ontologies. While existing initiatives such as IfcOWL simply map every IFC entity—including relationship objects like IfcRelAggregates, IfcRelConnects, and IfcRelDefinesByProperties—into OWL classes, this approach treats relationships as first‑class objects rather than as semantic links between domain entities. Consequently, RDF graphs become bloated with unnecessary intermediate nodes, query performance suffers, and the true meaning of the relationships is obscured.

The authors propose a two‑fold adaptation that respects both the object‑oriented constraints of the IFC schema and the need for compact STEP‑based files. First, they perform a syntactic transformation that maps IFC entities to OWL classes and IFC attributes to OWL data properties, preserving type information, units, and enumeration constraints. Second, and more importantly, they introduce a semantic transformation: every subclass of IfcRelationship is reinterpreted as an OWL object property. For each relationship, the “relating” and “related” roles are identified, and corresponding forward and inverse properties (e.g., hasPart ↔ partOf) are declared with explicit domain and range axioms. Cardinality restrictions (someValuesFrom, maxCardinality) are used to capture the original IFC multiplicity rules, and any auxiliary metadata (connection time, reason, etc.) is retained as OWL annotations.

To keep the resulting ontology lightweight, the authors design a mapping engine that eliminates empty relationship instances and collapses many‑to‑many connections into single triples wherever possible. The generated ontology conforms to the OWL 2 QL profile, ensuring that it can be stored and queried efficiently in large‑scale triple stores.

The experimental evaluation uses two real‑world IFC datasets: a modest residential building and a complex hospital design. Four metrics are measured: (1) total number of RDF triples, (2) file size, (3) SPARQL query execution time for typical BIM queries (e.g., “find all components that belong to a given assembly”), and (4) reasoning time for consistency checking with an OWL DL reasoner. Compared with the baseline IfcOWL representation, the proposed model reduces the triple count by an average of 27 % and the file size by 22 %. More strikingly, query execution times drop by 35 % to 48 % for multi‑join queries, demonstrating that the property‑based representation enables more direct graph traversal. Consistency checks show virtually no logical contradictions, confirming that the semantic mapping preserves the integrity of the original IFC constraints.

The discussion highlights several broader implications. By turning relationships into properties, the ontology becomes more amenable to downstream semantic technologies such as knowledge graphs, rule‑based reasoning, and machine‑learning pipelines that rely on clear predicate semantics. The use of OWL 2 QL also facilitates integration with existing BIM databases and web‑based services, promoting real‑time data exchange across the construction lifecycle. Moreover, the approach aligns with the goal of STEP file optimization, as fewer intermediate nodes translate into smaller serialized files and faster network transmission.

In conclusion, the paper delivers a concrete, standards‑compliant methodology for semantically adapting IFC model relations into OWL properties. The solution not only improves query performance and storage efficiency but also enhances the expressive power of BIM ontologies, paving the way for richer semantic applications in design, construction, and facility management. Future work is outlined to include dynamic synchronization of the ontology with live BIM updates and the integration of sensor data streams for operational building management.