Impact of occupancy behavior on building energy efficiency: What’s next in detection and monitoring technologies?

Abstract

Buildings account for a substantial portion of global energy consumption, and research indicates that occupant behavior can significantly influence energy use and building performance. This study provides a comprehensive review of recent progress in occupancy detection and monitoring technologies, highlighting how advanced methods can facilitate more accurate, occupant-driven energy management. Traditional sensor-based techniques such as CO₂ concentration monitoring, passive infrared (PIR) sensors, radio frequency (RF) signals, and indirectly, smart meter data are examined alongside more innovative, vision-based approaches incorporating deep learning and computer vision. Particular attention is paid to data-driven methods, including probabilistic models such as Hidden Markov Models (HMMs), classical machine learning algorithms such as Support Vector Machines (SVMs) and K-Nearest Neighbors (KNN), and deep learning architectures such as Convolutional Neural Networks (CNNs), all of which have demonstrated high accuracy in both laboratory and real-world settings. Emerging transformer-based fusion architectures and vision-language models (VLMs) are also discussed, highlighting their potential for capturing complex spatial-temporal occupancy patterns and enabling multimodal, interpretable occupancy detection. Despite their potential, numerous challenges remain. Privacy, data security, and user acceptance concerns must be addressed to ensure broad adoption; there is also a recognized need to improve the reliability of detection under varying environmental conditions. Personalization and adaptability emerge as key themes, particularly in multi-occupant contexts, while multi-sensor data fusion promises to enhance detection stability and reduce false positives. Finally, economic feasibility considerations such as installation costs and associated energy savings achieved through occupancy-driven heating, ventilation and air-conditioning (HVAC ) optimization are crucial for large-scale implementation. By synthesizing current methods, identifying research gaps, and proposing future directions, this review offers guidance for researchers and practitioners aiming to develop smart, occupant-centric building systems that balance energy efficiency, comfort, and privacy.