Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Indonesian Journal of Statistics and Its Applications
Hendra Rahmawan
School of Data Science, Mathematics and Informatics, IPB University, Indonesia
Computer Science & IT Mathematics Other

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Spatio-Temporal Clustering of PM2.5 Estimation in Jakarta Lutfiah Nursabiliyanti; Imas Sukaesih Sitanggang; Hendra Rahmawan; Muhammad Asyhar Agmalaro; Nor Azura Husin
Indonesian Journal of Statistics and Applications Vol 10 No 1 (2026): Vol 10 Issue 1 June 2026
Publisher : Statistics and Data Science Program Study, SSMI, IPB University, in collaboration with the Forum Pendidikan Tinggi Statistika Indonesia (FORSTAT) and the Ikatan Statistisi Indonesia (ISI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29244/ijsa.v10i1p79-93

Abstract

PM2.5 has negative impacts on human health because it can penetrate the alveoli of the lungs. This study aims to develop a PM2.5 estimation model for Jakarta based on Himawari AOD (Aerosol Optical Depth) and weather data from 2022 to 2024, utilizing a Random Forest Regressor and clustering with ST-DBSCAN. The data used in this study were PM2.5 measurements from eight air quality monitoring stations, recorded on the Jakarta Low Emissions website, and AOD Level 2 data from the Himawari-8 and Himawari-9 satellites, with a spatial resolution of 0.05 degrees, obtained from the Japan Aerospace Exploration Agency (JAXA) website. The results showed that the best PM2.5 estimation model was obtained with R2 = 0.63 and MAE = 8.035. Smoothing techniques for PM2.5 data have also been shown to improve model performance. Furthermore, based on the feature importance of the best PM2.5 estimation model, the Himawari AOD data are considered to have a less significant contribution to the PM2.5 estimation model. Clustering using ST-DBSCAN was successfully implemented on PM2.5 data for the 2022-2024 period, divided into 12 sub-datasets based on the seasons per year: the rainy season (DJF), the transition from rainy to dry (MAM), the dry season (JJA), and the transition from dry to rainy (SON). The best clustering result yielded a Silhouette coefficient of 0.75 on the 2023 rainy season (December–February) dataset. Three clusters were formed, consisting of 2094, 1458, and 878 data points, along with 45 noise points. The average PM2.5 levels in each cluster were 70.12 µg/m³, 62.71 µg/m³, and 52.18 µg/m³, respectively. The results of this study are expected to benefit other stakeholders involved in air pollution control in Jakarta.
Co-Authors Imas Sukaesih Sitanggang Lutfiah Nursabiliyanti Muhammad Asyhar Agmalaro Nor Azura Husin