Enhancing Durian Cultivation Efficiency Through Data-Driven Smart Farming Using Cluster Analysis and Machine Learning

This study explores the application of k-means clustering, combined with the elbow and silhouette methods, to analyze durian farm yields and production areas in Eastern Thailand from 2012 to 2023. The aim is to identify optimal farming practices and land use patterns to enhance productivity and sustainability. Using data on yield and area of production, the analysis reveals distinct clusters representing different farming characteristics. The evaluation metrics, including the Davies-Bouldin Index and Dunn Index, indicate that the Elbow method generally provides better defined clusters, although the Silhouette method occasionally shows superior clustering quality. The results show significant shifts in cluster centroids over the years, reflecting changes in smart farming. These insights suggest targeted interventions to optimize resource allocation and improve farm management.