Despite advances in building information modeling (BIM)–Internet of Things (IoT) integration, existing facility management workflows remain fragmented because no ontology framework currently unifies building models, IoT telemetry, manufacturer specifications and maintenance service data into a coherent structure. This semantic gap creates critical inefficiencies, decision bottlenecks and significant operational costs. To address this, the study develops and validates a unified semantic framework designed to replace disconnected, manual workflows with an automated, queryable and machine-interpretable knowledge infrastructure.
The study introduces a modular “hub-and-spoke” semantic architecture developed using the NeOn methodology. It comprises five domain-specific ontologies (BIMSO (core), Product Manufacturer Data Ontology (BIMMO), Internet of Things Ontology (IOTO), Facility Management and Maintenance Ontology (FMMO) and Maintenance Service Provider Ontology (MSPO)) interconnected via a shared upper ontology aligned with the UNIFORMAT II classification. The framework was validated on a real-world elevator system (ELEVATOR-EL001) in the Persian Gulf Complex. Competency questions were formalized as SPARQL queries and executed using Apache Jena Fuseki to assess semantic coherence, retrieval accuracy and time efficiency.
The framework's efficacy was validated on a real-world elevator system in a large-scale commercial complex. Cross-domain information retrieval tasks that took expert personnel 88–97 minutes to complete manually were successfully executed in under 2.2 minutes using the ontology, representing a time reduction of approximately 98% (over 40x faster). A paired t-test confirmed this gain is statistically significant (p < 0.001), with all query results verified for accuracy and semantic coherence. Beyond efficiency, the findings demonstrate the practical potential of unified ontologies to serve as the semantic backbone for scalable digital twin (DT), ultimately reducing operational risk and enabling real-time facility intelligence in complex building environments.
The framework reduces manual data aggregation efforts, improves maintenance response times and supports data-driven lifecycle management. Facility managers can leverage integrated insights for cost savings and operational efficiency.
This work presents a holistic, four-pillar integration model (unifying asset, operational, procedural and service knowledge) that, to the best of the authors' knowledge, for the first time, formally integrates manufacturer specifications and maintenance service contracts with BIM and IoT data into a single knowledge graph. By moving beyond the limited pairwise integrations of prior research, this approach provides a reusable and scalable foundation for creating intelligent digital twins and data-driven facility operations.
