How Can We End the Data ‘Silos’ in Smart Buildings?

The Problem of Disconnected Intelligence

By Antonia Egli and Radhika Deorukhkar (Dublin City University) and edited by Omar Doukari, Boubacar Seck, David Greenwood, Haibo Feng, and Mohamad Kassem

Imagine a high-tech university campus or a modern corporate office building. It’s packed with intelligence: smart lights adjusting to daylight, sensors tracking occupancy, and HVAC systems constantly regulating temperature. Yet, for many building managers, this smart infrastructure is more like a collection of talented individuals who refuse to talk to each other. The data from the lighting system sits in one digital silo, the energy meters in another, and the humidity sensors in a third. This lack of interoperability and the ability for different systems to exchange and use data is the silent bottleneck stifling the true potential of smart buildings. The study “Towards an Interoperable Approach for Modelling and Managing Smart Building Data: The Case of the CESI Smart Building Demonstrator”, conducted at Northumbria University, dives into this core challenge, exploring how a standardised, unified approach can make all that valuable data work together for a more efficient and sustainable built environment.

What Is the Interoperable Data Framework Study About?

This study addresses the critical need for a common language in the digital built environment. The research question focused on developing an interoperable approach for modelling and managing the complex, diverse data streams generated by building technology. The authors developed a robust framework using a concept called ontology modelling. An ontology is essentially a map that formally defines all the entities in a building (like rooms, fans, and sensors) and the precise relationships between them, all in a format machines can easily understand.

The framework was rigorously tested using the real-world complexities of the CESI Smart Building Demonstrator at the CESI Campus in Nanterre, Paris. The framework was rigorously tested on the Smart Building Demonstrator in Nanterre, Paris, access to which was kindly provided by CESI École d’Ingenieurs. This live testbed included data from its Building Energy Management System (BEMS), various HVAC components, electrical meters, and external weather stations. By applying the framework to this heterogeneous mix of data, the research aimed to validate a pathway toward truly integrated and intelligent building operations.

What Are the Core Findings of the Research?

The study confirmed that an ontology-based approach is the most effective way to unify diverse data sources, moving beyond simple data collection to create meaningful, integrated intelligence.

• A Unified Data Language: The framework successfully adapted open-source ontological standards, including the Building Topology Ontology (BOT) and the Brick Schema. These schemas were customised to create a single, unified data model for the entire CESI Demonstrator, successfully translating the different ‘languages’ of all the building’s systems into one consistent dictionary.
• Enabling Semantic Queries: Crucially, the unified model allows for semantic queries. Instead of needing to know the specific ID of a temperature sensor, a user or application can ask a query based on meaning, such as “Show the energy consumption of all heat-producing appliances on the second floor.” This level of intelligence facilitates advanced automation.
• The Foundation for Digital Twins: This interoperable framework provides the necessary, structured data foundation required to create a functional Digital Twin. A Digital Twin is a virtual replica of the physical building, allowing managers to run simulations, predict maintenance needs, and model energy consumption scenarios with high accuracy before implementing physical changes.
• Standardised Data Exchange: The study recommended using standardised protocols, such as JSON-LD and GeoJSON, for data representation and exchange. This ensures the data is not only understood internally but can also be easily shared with external applications and services, expanding its value dramatically.

How Does This Research Impact Industry and Sustainability?

These findings carry significant weight for industry, policy, and, critically, for achieving sustainability targets. When building data can talk to itself, new applications become possible, leading to tangible real-world benefits.

For industry professionals, the framework offers a roadmap to overcome vendor lock-in and system integration costs, two major hurdles in smart building adoption. By using open standards like BOT and Brick Schema, companies can mix and match technologies, ensuring their data remains accessible and useful regardless of the hardware brand.

This is particularly important in the context of global, and especially EU, sustainability goals. Smart buildings are essential tools for reducing operational carbon emissions and energy waste. The ability to cross-reference occupancy data from one sensor with HVAC performance from another, for instance, leads to real-time adjustments that can improve energy efficiency by up to 30% or more.

Real-World Scenario: Imagine a university using this framework. Instead of a monthly audit, the system constantly monitors when a lecture hall’s heating and ventilation were running while the occupancy sensors reported zero people. The system immediately flags this as a fault or inefficiency, allowing for immediate correction. This proactive fault detection and predictive maintenance drastically cut energy bills and extends the lifespan of expensive equipment. By providing a clear, tested method for data unification, the research accelerates the transition toward fully optimised and green infrastructure.

What Is the Key Takeaway and What Comes Next?

The journey towards truly intelligent, sustainable buildings begins with breaking down the digital walls that separate their data. This research provides a robust, tested, and ontology-driven approach that moves beyond simple connectivity to deliver semantic intelligence. The next step is to see this standardised framework adopted widely across the built environment, facilitating the development of a new generation of smart applications. The findings show that a unified data model is no longer a luxury, it’s the essential blueprint for a smarter, greener future.

Reference:

Doukari, O., Seck, B., Greenwood, D., Feng, H. and Kassem, M. (2022) ‘Towards an interoperable approach for modelling and managing Smart Building Data: The case of the CESI Smart Building Demonstrator’, Buildings, 12(3), p. 362. doi:10.3390/buildings12030362.