Analisis Keberagaman Input Produksi Usahatani Tembakau Kabupaten Bojonegoro Dengan Pendekatan Multivariat Cluster

The plantation sector plays a strategic role in the national economy, with tobacco serving as a key commodity that employs a significant number of workers. This is evidenced by the tobacco industry’s ability to employ an average of 47,840 people. Bojonegoro Regency is` one of the largest tobacco producers in East Java; however, its productivity remains volatile due to variations in farmers’ use of production inputs. To date, agricultural policies have tended to be general in nature and have not taken into account the differentiation in farming practices, thereby creating a significant disparity in the use of production inputs in the field. This study aims to cluster tobacco farmers based on their patterns of production input use using a Multivariate Cluster Analysis approach via the K-Means algorithm. The method used is quantitative descriptive analysis through a survey of 100 respondents selected via simple random sampling across six major production sub districts. The selection of the K-Means clustering algorithm is based on its effectiveness in classifying large datasets into homogeneous groups to identify farming behavior patterns at the micro level. The variables analyzed included land area, seeds, fertilizers, labor, pesticides, and capital, with cluster validity validated using the Elbow method and Silhouette Coefficient. The majority of farmers in Bojonegoro are of productive age with 16–25 years of farming experience; however, elementary school education dominates, and land ownership is limited (≤0.5 ha). The study identified two farmer clusters: Cluster 1 (22% of farmers) characterized by large-scale farming operations, and Cluster 2 (78% of farmers) characterized by small-scale operations. These findings confirm that strategies to increase tobacco productivity in Bojonegoro cannot be standardized. Future policies must be tailored and aligned with the specific characteristics of each cluster.