Developing functional digital twins for smart buildings requires accurate, real-time indoor localization tightly linked to the building’s spatial and semantic structure. However, most Indoor Positioning Systems (IPS) lack integration with Building Information Modeling (BIM), which limits 3D visualization and synchronized spatial management. This study proposes a BIM-driven hybrid localization framework that combines Wi-Fi and Bluetooth Low Energy (BLE) Received Signal Strength (RSS) data with BIM to create a unified data foundation for real-time 3D indoor positioning. The experimental system was implemented on the fourth floor of the Faculty of Geography at the University of Tehran, where a BIM model provided geometric and semantic references (IfcSpace, IfcStorey). RSS measurements from 35 reference points and seven transmitters were processed using two models: (1) Fingerprinting, implemented with a Multilayer Perceptron (MLP), and (2) Trilateration, based on a log-distance path-loss model. While Fingerprinting achieved high spatial accuracy (RMSE ≈ 0.40 m) and Trilateration showed larger positioning deviations (RMSE ≈ 2.38 m), a hybrid adaptive weighting strategy combined their outputs to obtain over 95% accuracy within one meter. The results confirm that BIM linkage enhances both positional precision and semantic consistency, enabling accurate mapping of user trajectories within a 3D building environment. The system operates through a three-tier workflow consisting of a mobile client, processing server, and BIM visualization layer, utilizing IFC and GeoJSON data formats. This framework provides a scalable foundation for digital-twin applications, with future extensions anticipated in multi-floor tracking, IoT data integration, emergency routing, and occupant flow analysis.