Clustering of Regencies and Cities in West Java Province Based on Horticultural Indicators Using the K-Means Method

Abstract

National food security largely depends on the capacity of domestic production. However, Indonesia continues to rely on food imports, including horticultural products. West Java Province, as one of the country’s main food-producing regions, possesses diverse geographical conditions that support the cultivation of various commodities and therefore becomes the focus of this study. This research aims to classify the 27 districts and cities of West Java Province based on horticultural indicators in order to identify spatial patterns and development potential. The study employed secondary data from the Central Bureau of Statistics (BPS) of West Java and World Climate for 2023, including horticultural production (ornamental plants, bio-pharmaca, vegetables, and fruits), annual average rainfall, and temperature. The analysis used the K-Means clustering method, with the Silhouette Index as an evaluation measure to determine the optimal number of clusters. Results indicate that three clusters provided the best accuracy. Cluster 1 (e.g., Cianjur and Bandung Regencies) consists of areas with high horticultural production, low temperatures, and moderate rainfall. Cluster 2 (e.g., Bogor and Sukabumi Regencies) represents regions with low production, moderate temperature, and high rainfall. Cluster 3 (e.g., Cirebon and Indramayu Regencies) includes areas with moderate production, high temperature, and low rainfall. The findings provide a foundation for local and national governments to design targeted horticultural development strategies that enhance productivity, improve farmers’ welfare, and support sustainable food security.