Development of a Predictive IoT-Based System for Real-Time Air Quality Monitoring and Route Optimization

Sri Lanka faces significant air pollution challenges, primarily from vehicular emissions, posing serious public health risks and impacting the quality of life. To address this, we developed a real-time air quality monitoring system that visualizes pollution data through an interactive heatmap. Our system integrates gas sensors that measure methane (CH<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</inf>), carbon monoxide (CO), carbon dioxide (CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>), sulfur dioxide (SO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>), nitrogen dioxide (NO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>), and particulate matter (PM.), connected to an ESP32 board and Wi-Fi module, and strategically deployed around smart buildings. The collected data is transmitted to a central hub using IoT technology and visualized on a mobile app developed with Flutter, featuring a color-coded scheme to represent pollution levels. This allows users to easily identify high-risk areas and select cleaner routes for travel. In testing, we focused on predicting PM. concentrations in Colombo using machine learning models, including Random Forest, XGBoost, and Prophet. Random Forest excelled in PM. prediction, while Prophet was highly effective for CO, demonstrating machine learning's potential in air quality forecasting. The system also includes a MIMI Air device that collects, processes, and transmits air quality data every 30 seconds, alerting users in real-time when pollutants exceed WHO thresholds, providing a cost-effective, reliable solution for public health.