Real-Time BIM–IoT Synchronization for Operational Digital Twins: The HPC4AI Data Center Case Study

Authors: Viviana Vaccaro, Lavinia Chiara Tagliabue, Robert Birke.

Abstract: Digital Twins (DTs) are emerging as a key enabling technology for the operational management of mission-critical infrastructures, where real-time integration between physical systems, Internet of Things (IoT) monitoring, and Building Information Modelling (BIM) can support optimization, predictive control, and sustainability. In this context, the University of Turin’s Computer Science Department is developing a DT of its HPC4AI data center, aimed at improving environmental monitoring, operational awareness, and energy-efficient management. The BIM model, originally generated from as-built documentation, was enriched to act as the geometric and semantic backbone of the DT. To create a live connection with the physical infrastructure, rack-level probes and environmental sensors are linked to BIM elements through dedicated shared parameters; this mapping is operationalized through a workflow based on Dynamo and custom Python scripts, which inject time-series measurements into the model as dynamic operational attributes rather than static annotations. Sensors inject timestamped measurements into a timeseries DB based on InfluxDB. Python scripts embedded in Dynamo periodically query InfluxDB via authenticated HTTP requests, parse the retrieved JSON data, and update the corresponding Revit parameters in real time, transforming the BIM model from a static geometric repository into a dynamic information system capable of reflecting the actual operational state of the data center. This process enables the representation of the current microclimatic state directly within the BIM environment. A dedicated alert engine evaluates incoming values against predefined thresholds and generates diagnostic flags and historical alert records. Graphical overrides highlight critical sensors through colour-coded states, allowing immediate spatial localization of anomalies. The methodology follows recent research trends demonstrating that BIM–IoT integration is essential for Facility Management- oriented DTs, enabling buildings to evolve into continuously updated, data-driven systems rather than fixed models. The paper discusses the system architecture, data synchronization strategy, and scalability for integrating additional sensors and control loops. Preliminary results show that the BIM–IoT bridge enhances situational awareness and lays the groundwork for predictive control and energy optimization of the data center infrastructure.

Keywords: Digital Twin, BIM–IoT Integration, Real-Time Data Synchronization, Building Management System, Energy Efficiency, Facility Management