Improving Health Through Indoor Environmental Quality Monitoring: A Review of Data-Driven Models and Smart Sensor Innovations

Research Team in Thermal and Applied Thermodynamics (2.T.A.), Mechanics, EnergyEfficiency and Renewable Energies Laboratory (L.M.3.E.R.), Department Engineering Sciences, Faculty of Sciences and Techniques Errachidia, Moulay Ismaïl University of Meknès, Errachidia, Morocco

^* Corresponding author: rachidkidari@gmailcom

Abstract

An important factor affecting building inhabitants’ comfort, well-being, and productivity is the quality of the indoor environment. There is a lot of promise in using artificial intelligence to manage environmental quality. AI offers a more effective and proactive method of improving indoor air quality and occupant well-being by predicting, monitoring, and regulating thermal comfort levels and lowering indoor pollution. The present study reviews recent scientific work on monitoring and improving indoor environmental quality (IEQ), focusing on the use of statistical learning models and smart sensor technology. Machine learning has been shown to effectively detect office occupancy using environmental measurements, improving energy efficiency and occupant comfort. Other research has successfully reconstructed indoor temperature profiles, essential for optimizing heating, ventilation and air-conditioning systems. Comprehensive reviews of air quality modeling in urban environments focus on the integration of advanced modeling techniques into urban planning. Studies on smart sensors for real-time monitoring of indoor air quality (IAQ) in various types of buildings demonstrate their potential for improving IAQ and thermal comfort. These studies underline the importance of data-driven approaches and intelligent systems in meeting the challenges of indoor environmental quality management. Future research should focus on integrating these technologies into intelligent building systems to improve energy efficiency, air quality and occupant comfort. Numerous cutting-edge deep learning techniques, including convolutional neural networks (CNNs), long short-term memory networks (LSTMs), decision trees (DTs), support vector machines (SVMs), artificial neural networks (ANNs), and deep neural networks (DNNs), are incorporated into the hybrid framework. Combining these methods improves the framework’s capacity to precisely process and examine intricate patterns of data.