Modelling and predicting classroom indoor air quality using machine learning algorithms

Background: Schools are a critical environment for children as they spend a significant amount of time in classrooms. Previous studies in school classrooms have revealed inadequate levels of ventilation and poor air quality. However, the cost of monitoring equipment and labour involved in detailed air quality monitoring make it challenging to quantify and manage indoor air quality (IAQ) in classrooms. The goal of this research is to use machine learning algorithms to analyse and predict IAQ, using minimal monitored data, thereby providing advance warnings to avoid peak air pollution concentrations in the classroom.

Method: This research will collect indoor and outdoor air pollution concentrations, including particulate matter (PM), carbon dioxide (CO₂), carbon monoxide (CO), nitrogen dioxides (NO2), ozone (O₃), and total volatile organic compounds (TVOCs) in selected schools that represent different exposure conditions. Data on classroom occupancy and activities as well as door window opening patterns will also be recorded. These data will be used to train machine learning models, such as artificial neural networks (ANNs), to predict indoor air pollutant concentrations. Machine learning, particularly through models like ANNs, excels at identifying complex patterns and can capture nonlinear relationships. These algorithms also can quantify the impact of various factors, such as ambient factors, building characteristics, and human activities, on IAQ.

Outcomes: The results of this research will include models that can predict real-time classroom air quality. This model will also identify key factors that contribute to poor air quality, such as high outdoor pollution levels, or specific classroom activities. Based on the predictive results, this research will assist in developing actionable recommendations for schools to minimize students’ exposure to harmful pollutants.

Impact: This research will offer models that serves as a reference for developing guidelines for school environments, which can be adopted by educational institutions, policymakers, and public health authorities. This will help to ensure good learning environments that can improve the health and academic performance of children.