Sensing of Particulate Matter (PM 2.5 and PM 10) in the Air of Tier 1, Tier 2, and Tier 3 Cities in India Using EVDHM-ARIMA Hybrid Model

Air pollution is now universally recognized as a public health and social hazard. Particulate matter (PM) is a significant air pollutant composed of liquid droplets such as acids, organic chemicals, metals, dirt, and dust. The concentration of PM in the air leads to numerous clinical symptoms of pulmonary and cardiovascular disorders. In many cases, respiratory diseases caused by the high concentration of PM has led to death in both human and animal. Hence, forecasting and sensing the PM concentration, specifically PM 2.5 and PM 10 in the air, turns out to be crucial. Such a prediction will help both the public and state administration of the region to take precautionary measures. Few studies have attempted to anticipate PM 2.5 in a single region, but their results have shown low efficiency. In this letter, a hybrid eigenvalue decomposition of the Hankel matrix (EVDHM) along with the autoregressive integrated moving average (ARIMA) based model has been used to construct a highly efficient model. To gain a comprehensive understanding of the created EVDHM and ARIMA-based model, three cities in India have been chosen. Three cities (Kolkata, Siliguri, and Haldia) have been selected as per the tier-wise classification of cities by the Government of India based on population. From 2021 to 2023, the daily PM 2.5 and PM 10 concentrations have been recorded. After training and testing the EVDHM-ARIMA-based model, it is compared with some well-performed models from the literature. The created EVDHM-ARIMA model is highly efficient and robust for practical applications.