Forecasting the air pollution index: a case study in Shah Alam, Selangor / Nur Hafiraniza Bakhtiar ... [et al.]

The World Health Organization (WHO) defines air pollution as any chemical, physical, or biological agent that tampers with the atmosphere's natural characteristics and contaminates either the indoor or outdoor environment. The evaluation of air pollution can be done by using air pollution prediction. When air pollution levels are high, it can notify and warn the public while assisting the management of many different chemical compounds through policy. The objective of this study is to find the best forecasting model for the air pollution index (API). This study also attempts to predict the monthly mean concentration of the API in Shah Alam for 2023 by using the time series model. To achieve the objectives, the Box-Jenkins Methodology and Univariate Techniques were used. This study examines the API using Holt’s Method, Double Exponential Smoothing Technique, and ARIMA models. Based on the smallest value of root mean squared error (RMSE) and mean absolute error (MAE), it shows that the most adequate model for the API for this period is the ARIMA model. Air quality forecasting is reliable and effective in controlling the composition of air pollution. With the ability to forecast the mean concentration of the Air Pollution Index, these findings could aid the Department of Environment in analyzing the substances that contribute to air pollution. Additionally, this information could help reduce the incidence of air pollution-related diseases among Malaysians.