Researchers associated with the NextAIRE consortium have contributed to a scientific publication in Atmosphere (MDPI), exploring how digital twin technology, artificial intelligence and low-cost sensors can improve indoor air quality monitoring and support smarter air purification strategies.
The paper, “Hybrid Digital Twin Framework for Real-Time Indoor Air Quality Monitoring and Filtration Optimization,” was published on 10 February 2026.
The study brings together researchers from several institutions, including Mario Lovrić and Nikolina Račić, whose research is supported by NextAIRE.

Combining AI, Sensors and Digital Twin Technology
The researchers developed a hybrid digital twin framework that combines real-time sensor measurements, physics-based simulations and machine learning models.
The system was evaluated during a 26-day experimental campaign in a residential environment near Zagreb, Croatia.
The research focused on monitoring indoor pollutants, including particulate matter (PM), volatile organic compounds (VOCs) and carbon dioxide (CO₂).
The researchers also investigated how the positioning and operation of an air purifier could affect indoor air quality.
Key Research Findings
The study demonstrated that:
- Machine learning models can accurately predict several indoor air quality parameters.
- Linear regression models achieved particularly strong performance under limited-data conditions, with R² values between 0.97 and 0.99 for several parameters.
- Air purification was associated with reduced particulate matter concentrations under the experimental conditions.
- Digital twin technology can help visualise air pollution patterns and assess different air purification strategies.
Supporting Smarter and Healthier Indoor Environments
The findings highlight the potential of combining environmental monitoring, artificial intelligence and physical simulations to improve indoor air quality management.
Although the experimental evaluation was limited to a single residential room, the proposed framework demonstrates promising potential for future applications in smart buildings and indoor environmental management.
The research aligns with NextAIRE’s objectives of advancing AI-driven air quality monitoring, interdisciplinary collaboration and innovative environmental technologies.
Read the Full Publication
Title: Hybrid Digital Twin Framework for Real-Time Indoor Air Quality Monitoring and Filtration Optimization
Authors: Valentino Petrić, Dejan Strbad, Nikolina Račić, Tareq Hussein, Simonas Kecorius, Francesco Mureddu and Mario Lovrić
Journal: Atmosphere (MDPI), 2026, 17(2), 184
Full article: https://www.mdpi.com/2073-4433/17/2/184