Predicting Air Quality in New Delhi using an AUTO-ARIMA Machine Learning Algorithm

The increase of air pollution levels in urban areas especially in densely populated cities like New Delhi has made air quality a major public health concern. The National Institute of Health in Delhi has found that hospital visits for respiratory diseases increase during New Delhi's winter and autumn seasons alongside the peak air pollution during winter. Research using Generalized Additive Models (GAM) for single and combined pollutant situations has shown that respiratory disease cases rise among urban residents following short-term exposure to PM10 and SO2 air pollutants. In this study, data was collected from Kaggle where monitoring stations were placed to gather pollution data from multiple locations around Delhi. This data covers the period from 2020 to 2023. The pollutant data was given in terms of concentrations that subsequently were converted to air quality index AQI. This study analyzes and predicts New Delhi's AQI from comprehensive pollution data that measures particulate matter (PM2.5 and PM10), nitrogen dioxide (NO2), sulfur dioxide (SO2), carbon dioxide (CO2), ozone (O3), and additional air pollutants from November 25, 2020, through January 2024. We constructed a three-year AQI predictive model through both statistical and machine learning methods applied in Google Colab using ARIMA modeling. The model was developed by splitting the data in 80:20 ratio where 80 percent data was used to train the model, and the remaining 20 percent was used as the test data. The model predicts future pollution levels by analyzing historical pollutant data together with environmental variables to map air quality patterns. Model predicted AQI values in next three years with 29.78 rmse values. The reason for model poor performance was the data itself that was not consistent with multiple trends. Sophisticated algorithms may require to improve the model accuracy. In general, this research provides Policymakers with essential information to develop effective air quality management strategies that will help reduce pollution effects in New Delhi.