Rice is one of the main commodities in Indonesia's agricultural sector that plays a crucial role in maintaining national food security. This study aims to cluster 38 provinces in Indonesia based on the similarity of rice production in order to understand the spatial variation of agricultural performance in different provinces. In this analysis, three clustering methods were used, namely K-Means, K-Medoids, and Fuzzy C-Means, by considering two main variables: the area of harvest and the amount of rice production in 2024, whose data were sourced from the Central Bureau of Statistics. Evaluation of cluster quality was conducted using the Davies-Bouldin Index (DBI). The results showed that the K-Means method produced the most optimal clustering with the lowest DBI value of 0.276 at the number of clusters , compared to K-Medoids (0.279 at ) and Fuzzy C-Means (0.285 at ). The clusters formed show a clear separation between provinces with high and low production levels. Provinces with high agricultural intensification and a large contribution to production belong to the main cluster, while areas with limited resources and low production form a separate cluster. Several other clusters reflect medium to high production characteristics with varying development potential. This finding reflects the diversity of agricultural conditions influenced by infrastructure, intensification, and geographical and climatic factors.
Mujahidin et al. (Thu,) studied this question.