Integrating ventilation assessments with IoT monitoring to evaluate indoor air quality in elementary schools of San Luis Potosí, México

Indoor air quality (IAQ) is often neglected in schools of the Global South. Carbon dioxide (CO 2 ) and fine particulate matter (PM 2.5 ) serve as key indicators of ventilation and exposure risk. This study employed a two-pronged approach to evaluate IAQ in elementary schools of the San Luis Potosí Metropolitan Area, Mexico. Ventilation rates were quantified in 106 spaces across 53 schools. Continuous IAQ monitoring (CO 2 , PM 2.5 , temperature, relative humidity) was conducted using low-cost Air Sentinel monitors in a convenience sample of 11 schools during cold and hot seasons. Machine learning models forecasted CO 2 concentrations one hour in advance, and the Wells-Riley model estimated airborne infection probability (AIP). IAQ was classified as poor, acceptable, and excellent in 1.3%, 31.5%, and 67.2% of classrooms, respectively. During the cold season, 97.7% of classrooms maintained CO 2 within 400–1000 ppm; in the hot season, 67.7% fell within this range and 1.2% exceeded 2000 ppm. The strongest CO 2 forecasting performance occurred in the hot season, while PM 2.5 forecasts failed in both seasons, AIP increased monotonically during the first two hours of class in both seasons. We conclude that integrating ventilation assessments with continuous IoT monitoring provides a comprehensive approach to characterizing IAQ in resource-constrained settings.