CLUSTERING OF REGENCIES/CITIES IN WEST JAVA PROVINCE BASED ON INDICATORS THAT AFFECT AIR QUALITY USING PRINCIPAL COMPONENT ANALYSIS AND K-MEANS CLUSTERING

Abstract

Air quality in Indonesia has significantly declined over the past two decades, transforming from one of the cleanest countries in 1998 to one of the twenty most polluted by 2016 due to a 171% increase in air particulate pollution concentrations. This study examines factors affecting air quality in West Java Province, one of Indonesia’s most populous regions, using Principal Component Analysis (PCA) and K-Means Clustering. The analysis includes 14 variables, such as population density, forest area, road length, and vehicle numbers. PCA was used to reduce data dimensions while retaining essential characteristics, identifying two principal components: population mobility and land infrastructure. The cluster analysis revealed two distinct groups: Cluster 1 includes 18 regencies/cities with lower population mobility and land infrastructure, indicating better air quality, while Cluster 2 consists of 9 regencies/cities with higher population mobility and land infrastructure, potentially reflecting worse air quality. Areas in Cluster 2 are concentrated near DKI Jakarta, Bandung, and the eastern border with Central Java, suggesting the influence of urbanization, industrial activities, and cross-border emissions. This study provides a spatial grouping of regencies/cities in West Java based on air quality indicators, offering insights for policymakers to target interventions more effectively. The findings emphasize the need for sustainable urban planning and stricter environmental regulations to address the growing air pollution challenge.