Forecasting the Air Quality Index (AQI) in Jakarta, Indonesia by Using a Linear Regression Model

Air quality conditions currently demands particular consideration, notably in Jakarta. As per the Air Quality Index (AQI) website, Jakarta ranks second globally for the poorest air quality, registering an AQI value of 170 (categorized as unhealthy). To address this challenge, forecasting emerges as...

Full description

Saved in:
Bibliographic Details
Main Author: Zidan Aqila Kamil, Mochamad
Format: Final Year Project
Language:English
Published: 2024
Subjects:
Online Access:http://utpedia.utp.edu.my/id/eprint/27001/1/21002801.pdf
http://utpedia.utp.edu.my/id/eprint/27001/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Air quality conditions currently demands particular consideration, notably in Jakarta. As per the Air Quality Index (AQI) website, Jakarta ranks second globally for the poorest air quality, registering an AQI value of 170 (categorized as unhealthy). To address this challenge, forecasting emerges as a potential solution. Forecasts serve as essential tools for policymakers and the public, enabling proactive measures to regulate and prevent future air quality issues. Among the methodologies available for forecasting, the Linear Regression model stands as one viable approach. The testing process was carried out using daily Index of Air Quality Standard (ISPU) DKI Jakarta data (1 March 2021 to 31 December 2021) obtained from the Satu Data Jakarta website, with 80% of the data as training data and 20% as test data. The parameters predicted by the Linear Regression model are the concentration values of the pollutants Particulate Matter 25 (PM25), Particulate Matter 10 (PM10), Sulphur Dioxide (SO2), Carbon Monoxide (CO), Ozone (O3), and Nitrogen Dioxide (NO2), with evaluation using the Mean Absolute Percentage Error (MAPE) and Root-Mean-Square Error (RMSE) metrics. Overall, the results of forecasting pollutant parameters using the Linear Regression model obtained good accuracy. Very accurate results (MAPE < 10%) were obtained by the SO2 parameter. Then accurate results (MAPE 11% - 20%) were obtained by the O3 parameter. The rest got fairly accurate results (MAPE 21% - 50%) obtained by the parameters PM2.5, PM10, CO and NO2. Apart from that, visualisation of forecasting results is presented in the form of a website, along with the Air Quality Index (AQI) value, parameter value, and AQI category.