Smart Building Technology and EPBD: How IoT Drives Compliance

Why Smart Buildings Are Central to EPBD Compliance

Buildings account for approximately 40% of total energy consumption and 36% of greenhouse gas emissions across the European Union. These numbers make the built environment one of the most significant levers available to policymakers pursuing decarbonization targets. The recognition of this fact is precisely what drove the European Commission to undertake a comprehensive overhaul of building energy regulation, culminating in the adoption of Directive 2024/1275 — the EPBD recast — in April 2024.

The recast does not merely tighten existing requirements. It fundamentally reframes how buildings are expected to operate by making smart building technology a core compliance mechanism rather than an optional enhancement. Articles 14 and 15 of the directive introduce mandatory requirements for Building Automation and Control Systems (BACS) and establish the Smart Readiness Indicator (SRI) as a standardized framework for assessing a building's technological capability. For the first time, the directive explicitly links a building's regulatory standing to its ability to monitor, analyze, and optimize its own energy performance through digital systems.

This shift has profound implications for building owners, property managers, facility operators, and the technology companies that serve them. Smart building technology — encompassing IoT sensors, edge computing, cloud analytics, and AI-driven optimization — is no longer a competitive differentiator. It is becoming a regulatory baseline. Buildings that lack the ability to continuously monitor energy consumption, respond to occupancy patterns, and communicate with the electrical grid will face compliance gaps that manual processes cannot close.

The convergence of regulatory pressure and technological maturity creates a clear imperative: invest in smart building infrastructure now, or face escalating compliance costs and potential penalties as national transposition deadlines take effect across EU member states starting in 2026.

EPBD Requirements for Building Automation (Art. 14-15)

The EPBD recast introduces two distinct but complementary frameworks that directly govern the use of smart technology in buildings: the BACS mandate under Article 14 and the Smart Readiness Indicator under Article 15.

Building Automation and Control Systems (BACS) — Article 14

Article 14 requires the installation of building automation and control systems in all non-residential buildings where the effective rated output of the heating, ventilation, or air conditioning systems exceeds 290 kW. This threshold was already in effect as of 2025. By 2030, the threshold drops to 70 kW, bringing a far larger population of commercial buildings under the mandate.

The directive specifies that BACS must be capable of:

• Continuously monitoring, logging, and analyzing energy use by major system (heating, cooling, ventilation, lighting, domestic hot water)
• Benchmarking the building's energy efficiency against internal baselines and external references
• Detecting losses in efficiency of technical building systems and informing the facility manager
• Allowing communication between interconnected technical building systems and other appliances inside the building
• Adjusting energy flows based on demand signals, occupancy, and external conditions

The European standard EN 15232 provides the classification framework for BACS energy performance impact, dividing systems into four classes:

• Class A — High energy performance building automation and control systems with comprehensive automated control, real-time optimization, and fault detection
• Class B — Advanced building automation with automated demand-based control for major systems
• Class C — Standard building automation with basic time-based scheduling and manual override
• Class D — Non-energy-efficient systems with no automated control

To meet the EPBD mandate, buildings must achieve at least Class B performance. Buildings operating at Class C or D must upgrade their automation systems or demonstrate equivalent performance through alternative measures. Studies based on EN 15232 suggest that moving from Class D to Class A can reduce thermal energy use by 30-50% in non-residential buildings.

Smart Readiness Indicator (SRI) — Article 15

Article 15 establishes the SRI as a voluntary but increasingly important rating tool. The SRI assesses a building's capacity to adapt its operation to the needs of occupants and the grid, and to improve its energy efficiency. While member states are not yet required to mandate SRI assessments, the indicator is expected to become a standard component of Energy Performance Certificates (EPCs) and a factor in property valuations. We will examine the SRI framework in detail in a subsequent section of this article.

IoT Sensor Types for EPBD Compliance

Meeting the BACS requirements and achieving a strong SRI score depends on deploying the right sensor infrastructure throughout a building. Each sensor type maps to specific compliance data requirements, and the combination of sensors determines the building's ability to monitor, analyze, and optimize its energy performance in real time.

Energy Meters

Sub-metering is the foundation of EPBD-compliant energy monitoring. Buildings need electricity sub-meters at the distribution board level to separate consumption by end use — lighting, HVAC, plug loads, elevators, and common areas. Gas meters track heating fuel consumption, while district heating sub-meters measure thermal energy delivered from external networks. The directive requires granular, time-stamped energy data, making smart meters with 15-minute or hourly interval logging essential.

HVAC Sensors

Heating, ventilation, and air conditioning systems typically account for 40-60% of a commercial building's energy consumption. HVAC-specific sensors include temperature sensors (supply air, return air, zone-level), humidity sensors (for dehumidification control and comfort management), airflow sensors (measuring duct velocity and volume for ventilation verification), and pressure sensors (monitoring filter condition and duct integrity). These sensors enable the demand-based control required for EN 15232 Class B or higher.

Occupancy Sensors

Occupancy-driven energy management is a core requirement for advanced building automation. Passive infrared (PIR) sensors detect presence for lighting and HVAC zone control. CO2-based occupancy estimation uses indoor carbon dioxide levels as a proxy for the number of occupants, enabling proportional ventilation control that saves energy while maintaining air quality. Combining PIR with CO2 sensing provides both binary presence detection and headcount estimation.

Water Meters

While water consumption is not the primary focus of the EPBD, smart water meters contribute to the building's overall resource efficiency profile and SRI score. Sub-metering domestic hot water separately from cold water enables identification of heat losses in distribution systems and supports whole-building resource optimization.

Indoor Air Quality Sensors

Indoor environmental quality is now part of the expanded EPC framework under the recast. Sensors measuring CO2 concentration (in parts per million), volatile organic compounds (VOCs), and particulate matter (PM2.5) provide the data needed for both compliance reporting and demand-controlled ventilation. Maintaining CO2 below 1000 ppm and PM2.5 below 25 µg/m³ aligns with WHO guidelines and supports occupant health claims.

Solar and Renewable Generation Monitors

Buildings with on-site renewable energy systems — primarily rooftop photovoltaic arrays — need generation monitoring to verify self-consumption rates and export volumes. Current transformers on inverter outputs, combined with irradiance sensors, enable performance ratio calculations that detect degradation or faults in the PV system. This data is critical for zero-emission building (ZEB) verification under the directive.

The following diagram illustrates how these sensor types are typically distributed across a building and connected to the analytics platform:

The Smart Readiness Indicator (SRI) Explained

The Smart Readiness Indicator is a standardized rating system that measures a building's ability to use information and communication technologies to adapt its operation to the needs of occupants and the grid, and to improve its energy and operational efficiency. Introduced by the EPBD recast under Article 15, the SRI provides a common language for comparing the technological maturity of buildings across the EU.

What the SRI Measures

The SRI evaluates buildings across three key impact criteria, also called functionality domains:

• Energy Efficiency — The building's ability to optimize energy consumption through automated control of HVAC, lighting, domestic hot water, envelope management (automated blinds, operable windows), and on-site renewable energy systems. This domain assesses whether the building can reduce energy waste without manual intervention.
• Response to Occupant Needs — The building's capacity to maintain and improve indoor environmental quality, including thermal comfort, visual comfort, indoor air quality, and acoustic performance. Systems that allow personalization — individual zone control, user-adjustable setpoints — score higher.
• Energy Flexibility — The building's readiness to participate in demand response programs, shift loads to off-peak periods, store energy (thermal or electrical), manage electric vehicle charging, and interact dynamically with the electrical grid. This domain is becoming increasingly important as EU energy markets evolve toward real-time pricing and distributed energy resources.

SRI Scoring Methodology

The SRI assessment evaluates individual technical building systems — heating, cooling, ventilation, lighting, dynamic envelope, electricity, electric vehicle charging, and monitoring and control — against a catalogue of smart-ready services. Each service is rated on a scale from Level 0 (no smart functionality) to Level 4 (fully optimized with predictive control and grid interaction). The scores are aggregated into percentage ratings for each of the three impact domains, and then combined into an overall SRI score.

A building with no automation might score 5-15% overall. A fully automated, AI-optimized building with grid interaction capabilities could score 85-100%. Most existing commercial buildings fall in the 20-50% range, indicating significant room for improvement through targeted technology upgrades.

How to Improve SRI Score Through Automation

Improving a building's SRI score is a systematic process that starts with an audit of existing systems against the SRI service catalogue. Common high-impact upgrades include:

• Installing demand-controlled ventilation with CO2 sensing (improves both Energy Efficiency and Occupant Comfort domains)
• Deploying occupancy-based lighting and HVAC zone control
• Adding automated shading systems (dynamic envelope management)
• Implementing a central building management system with open communication protocols (BACnet, Modbus, KNX)
• Enabling bidirectional EV charging infrastructure for vehicle-to-building energy flexibility
• Integrating battery storage with time-of-use optimization algorithms

Each upgrade moves specific services from lower to higher functionality levels. A well-planned technology roadmap can increase a building's SRI score by 20-40 percentage points within a single investment cycle, often with payback periods under five years thanks to the associated energy savings.

Data Architecture for Smart Building Compliance

Deploying sensors is only the first step. The data those sensors generate must be collected, processed, stored, and made accessible for both real-time operations and regulatory reporting. A well-designed data architecture is the connective tissue that turns hardware investments into compliance outcomes.

Edge Computing for Local Processing

Smart buildings generate enormous volumes of data — a single commercial building with comprehensive IoT instrumentation can produce millions of data points per day. Processing all of this data in the cloud introduces latency, bandwidth costs, and reliability risks. Edge computing addresses this by placing processing power at the building level.

Edge gateways aggregate data from sensors using protocols like MQTT, BACnet/IP, Modbus TCP, and LoRaWAN. They perform local filtering, aggregation, and rule-based actions (for example, shutting down HVAC in unoccupied zones) with sub-second response times. Only aggregated metrics and anomaly alerts are forwarded to the cloud, reducing bandwidth requirements by 80-90% compared to raw data transmission.

Cloud Analytics Platform

The cloud layer handles long-term storage, advanced analytics, machine learning model training, and multi-building portfolio management. Time-series databases are the natural fit for building sensor data. InfluxDB and TimescaleDB are the two most widely adopted options, both offering high write throughput, efficient compression of time-stamped data, and built-in downsampling for historical queries. A typical deployment stores raw 15-minute interval data for one year and downsampled hourly data for five or more years, aligning with EPC validity periods.

API Standards and Semantic Interoperability

One of the biggest challenges in smart building data management is the heterogeneity of systems. A single building might contain equipment from dozens of manufacturers, each with its own data model and naming conventions. Semantic data standards address this problem by providing a common vocabulary for describing building systems, sensors, and their relationships.

Brick Schema is an open-source ontology that defines a standardized set of tags and relationships for describing building equipment, sensors, and spaces. It enables queries like "find all temperature sensors on floor 3 that are associated with AHU-1" regardless of the underlying hardware vendor.

Project Haystack is a complementary tagging standard widely used in building automation, providing a lightweight metadata model for labeling and organizing operational building data.

Adopting one or both of these standards dramatically reduces the integration effort when onboarding new buildings or replacing equipment, and ensures that compliance reporting tools can access data consistently across the portfolio.

Data Governance and GDPR Considerations

Smart building data includes occupancy patterns, zone-level presence detection, and in some cases individual comfort preferences. Under GDPR, any data that can be linked to an identifiable individual — including desk-level occupancy in assigned workspaces — is personal data and must be handled accordingly. Building operators must implement data minimization (aggregate occupancy by zone rather than tracking individuals), purpose limitation (use occupancy data only for energy optimization and compliance), and appropriate retention policies. Privacy impact assessments should be conducted before deploying occupancy sensors in tenant-occupied spaces.

AI-Driven Building Optimization

Once the sensor infrastructure and data architecture are in place, artificial intelligence transforms raw building data into autonomous optimization. AI does not merely report on energy performance — it actively improves it, often achieving savings that exceed what manual tuning can accomplish.

HVAC Optimization

AI-driven HVAC control is the single highest-impact application of machine learning in buildings. By learning the thermal dynamics of a building — how quickly zones heat up, how weather affects cooling loads, how occupancy patterns vary by day and season — AI controllers can anticipate demand rather than merely reacting to it. Research and commercial deployments consistently demonstrate energy savings of 20-30% compared to rule-based BMS control, with some projects reporting savings up to 40% in buildings with high variability in occupancy and weather exposure.

Occupancy-Based Energy Management

AI models that fuse data from multiple sensor types — PIR, CO2, WiFi device counts, calendar systems — can predict occupancy 1-4 hours in advance with over 90% accuracy. This enables pre-conditioning of spaces before occupants arrive and early ramp-down when spaces are expected to empty, eliminating the energy waste of maintaining comfort conditions in unoccupied zones.

Predictive Maintenance

Machine learning models trained on vibration, temperature, pressure, and energy consumption data can detect equipment degradation weeks before failure occurs. Studies across commercial building portfolios show that predictive maintenance reduces unplanned equipment failures by 35-40% and extends equipment lifespan by 15-25%. For EPBD compliance, the benefit is direct: equipment operating at peak efficiency maintains the building's energy performance rating, while degraded equipment causes EPC drift toward lower ratings.

Fault Detection and Diagnostics (FDD)

AI-powered fault detection continuously compares actual system behavior against physics-based and data-driven models. Common faults detected include simultaneous heating and cooling (a pervasive issue in multi-zone VAV systems), stuck dampers and valves, sensor drift, economizer failures, and suboptimal scheduling. FDD systems can identify and diagnose faults that waste 10-30% of HVAC energy, often on the same day they begin. Without FDD, these faults can persist for months or years, silently degrading energy performance.

To learn more about how artificial intelligence can transform your building operations, explore SDA's AI & Automation capabilities.

How SDA Implements Smart Building Solutions

SDA brings together IoT engineering, cloud architecture, and AI expertise to deliver smart building solutions that are purpose-built for EPBD compliance. Our approach is pragmatic: start with a focused proof of concept, demonstrate measurable results, and scale from there.

Smart Community PoC

Our Smart Community PoC is a deployable reference implementation that includes ESG dashboards, IoT sensor integration, energy monitoring, and compliance reporting. The PoC covers the full stack — from sensor data ingestion through edge gateways to cloud analytics and user-facing dashboards. It is designed to be operational within two months, giving building owners a working compliance platform without the risk and timeline of a ground-up custom build.

The IoT and Smart Building feature of the Smart Community PoC includes pre-built integrations for common sensor types (energy meters, HVAC sensors, occupancy detectors, air quality monitors), a time-series data pipeline, and dashboard widgets for SRI-aligned reporting.

Rental Management PoC

For residential and mixed-use portfolios, our Rental Management PoC extends the smart building capabilities with tenant-facing features. The scale_beyond module includes HVAC control interfaces that allow tenants to adjust comfort settings within EPBD-compliant energy budgets, ensuring that individual comfort does not compromise building-wide energy targets. Automated compliance reporting per unit simplifies the administrative burden for landlords managing hundreds of rental properties.

Our Delivery Model

Every engagement begins with a sensor audit and data architecture assessment. We identify which sensors are already deployed, which gaps exist relative to BACS and SRI requirements, and design the minimum viable sensor network needed for compliance. From there, we build the data pipeline, deploy analytics, and iterate based on real operational data. The two-month PoC timeline includes sensor integration, dashboard development, and initial AI model training for HVAC optimization and fault detection.

Conclusion

The EPBD recast has made smart building technology a regulatory requirement, not a luxury. The mandatory BACS provisions under Article 14, the SRI framework under Article 15, and the expanded EPC data requirements collectively demand that buildings be equipped with comprehensive sensor networks, robust data architectures, and intelligent automation systems.

The technology to meet these requirements exists today. IoT sensors are mature, affordable, and available in wireless form factors that minimize installation disruption. Edge and cloud platforms handle the data engineering. AI delivers the optimization that turns compliance from a cost center into a source of operational savings — 20-30% energy reduction through HVAC optimization alone, with additional gains from predictive maintenance, fault detection, and occupancy-based management.

The buildings that act now will not only meet the letter of the EPBD but will position themselves for higher property valuations, lower operating costs, improved tenant satisfaction, and readiness for the increasingly dynamic energy markets of the next decade. The buildings that wait will face compressed timelines, higher costs, and the real risk of non-compliance penalties as national transposition deadlines arrive.

Smart building technology is the bridge between where the European building stock is today and where the EPBD requires it to be by 2050. The time to start building that bridge is now.