MAPPING FOOD SECURITY DIVERSITY IN SUMATRA: A K-MEANS CLUSTERING APPROACH

Abstract

Food security remains a strategic issue in Indonesia, particularly in Sumatra, which plays a crucial role as one of the country's major food-producing regions. Despite its substantial agricultural potential, significant disparities in food security conditions persist across provinces. This study aims to identify regional typologies of food security and socioeconomic conditions in Sumatra using the K-Means Clustering approach. The analysis employs five indicators, namely the Food Security Index (FSI), Normalized Difference Vegetation Index (NDVI), Night-Time Light (NTL), agricultural production, and poverty rate. Data from ten provinces in Sumatra during the period 2019–2024 were utilized. The optimal number of clusters was determined using the elbow method, which indicated that three clusters provided the most appropriate grouping structure. The results reveal substantial heterogeneity among provinces. Cluster 1 consists of Bengkulu, Jambi, and Riau, characterized by moderate food security, relatively high economic activity, and moderate agricultural production. Cluster 2 includes the Riau Islands and Bangka Belitung Islands, which exhibit low agricultural production but relatively stable food security and low poverty rates. Cluster 3 comprises Aceh, Lampung, West Sumatra, South Sumatra, and North Sumatra, functioning as major food-producing regions with varying socioeconomic vulnerabilities. The findings uncover a food security paradox in Sumatra, where high agricultural production does not necessarily correspond to stronger food security outcomes, while regions with limited production can maintain relatively stable food security through economic integration and food distribution networks. These results highlight the importance of adopting region-specific food security policies that consider not only production capacity but also socioeconomic and accessibility dimensions.