Influences of Human Presence on the Indoor Air Quality of Educational Institutions: Concurrent Multipollutant Sensing Approach

This study evaluates how human presence influences classroom air quality using an edge-intelligent, low-cost monitoring system [air quality monitoring system (AQMS)] that senses multiple pollutants and quantifies occupancy with an on-board camera and custom object-detection model. Over seven days in a laboratory, we varied air-conditioning and ventilation settings while the AQMS logged pollutant levels every minute. Spearman rank-order analysis showed strong positive correlations between occupancy and CO2, total volatile organic compounds (VOCs), and gas-sensor resistance, but no significant link with CO, NH3, or NO2. Particulate matter (PM2.5 and PM10) briefly fell as occupancy rose, yet this trend vanished when air-conditioning was active, indicating its filtration effect. Rooms without air-conditioning but with adequate ventilation maintained lower CO2 and VOC levels. PM findings were inconclusive because of concurrent inhalation effects. All occupancy data were privacy-preserving—images were processed on the device and never stored. Although demonstrated in a single room, the architecture is inherently scalable, and a multiroom urban–rural deployment is underway.