An Intelligent Geofenced Air Quality Monitoring System: Real-Time AQI Detection and Autonomous Location-Based Health Intervention Using Machine Learning

The rising level of air pollution in the urban environment triggers the need for intelligent air pollution monitoring systems and proactively protective health systems for the welfare of society as a whole. The paper presents the concept of the ‘Intelligent Geo fenced Air Quality Monitoring System’ that combines the strength of air pollution sensing using Internet-of-Things technology and health protective interventions with the power of machine learning analytics, as well as geographical-based health protective systems by using the Node MCU/Arduino platform with low-cost air pollution sensors such as MQ-135 & PM2.5 sensors. Advanced machine learning techniques are applied to bring about more analytical precision and predictive capability. The Random Forest models classify the state of air quality into safety categories, while the Long Short-Term Memory (LSTM) networks understand temporal dependencies in predicting AQI trends. GPS- enabled Geofencing, along with Haversine distance computation, identifies users within high-risk zones in real time. Once unhealthy conditions are detected, automated alerts are published through Firebase Cloud Messaging to mobile applications created on Android/Flutter systems. The proposed system contributes to a scalable, energy-efficient, cost- effective platform for smart air quality surveillance to enable preventive public health measures that support sustainable smart city ecosystems.