AI-Based Air Quality Prediction and Pollution Source Analysis System for Delhi NCR

Main Article Content

Vinay Lodha

Abstract

Delhi NCR is severely affected by pollution from Vehicles, industries, construction dust, and seasonal crop burning. Although data is available, it is dispersed across various systems and does not provide a clear picture of the primary cause of pollution or its potential effects. Decisions are often made only after the situation gets worse.


 


This paper describes an AI-based system for identifying pollution sources, predicting AQI levels, and supporting decision-making in Delhi NCR. The system gathers data from sensors, satellite sources, weather conditions  and traffic pattern in a single platform.


It utilizes Machine Learning models to identify pollution hotspots, major sources, and predict future AQI levels. The results are presented using a real-time dashboard that helps authorities to monitor the condition and take action accordingly. The system is aligned with initiatives by the Ministry of Environment and related pollution control bodies in India.


 


This platform is built using Python for Machine Learning, React.js for frontend of dashboard and cloud services for data management and storage. It supports role-based access, where users can view and use the system according to their responsibilities where administrators monitor data, analysts study trends, and policymakers make decisions. The system supports environment sustainability goals and help in proper pollution management

Article Details

Section

Articles

Author Biography

Vinay Lodha

 Department of Computer Science & Engineering

              Geetanjali Institute of Technical Studies (GITS)                                  

                                      Udaipur, India                                                                                         

References

SAFAR: reference for pollution app ui.

Daily scale air quality index forecasting using bidirectional recurrent neural networks: Uses BILSTM emphasizing pollutants like PM2.5/NO2.

Satellite Observations for Air Quality (NASA Terra/ MODIS): Explains MODIS for identifying pollution sources like fires and aerosols

MODIS Data Products for Aerosol and Pollution Monitoring:(NASA MODIS Web): (NASA MODIS Web): Official guide to MODIS AOD.

Air pollution prediction with machine learning case study of Indian cities:ML for hyperlocal routing.

AirNow.gov Interactive Map and Dashboard.

Real-time policy tool for urban AQI management and intervention monitoring