<![CDATA[Oil palm is one of the important plantation commodities that plays a significant role in Indonesia’s economy. PT. Suryabumi Agrolanggeng has historical oil palm production data covering production, rainfall, and fertilizer usage variables for the period 2019–2025. The main challenges are production fluctuations and the suboptimal utilization of historical data to identify production patterns and periods with different characteristics. This study aims to apply the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method to group data based on the similarity of their characteristics. The research stages include data collection, preprocessing, standardization using the Z-Score method, parameter determination using a K-Distance Graph, DBSCAN clustering, and evaluation using the Silhouette Score. The results show that the parameters Eps = 0.95 and MinPts = 6 produce three main clusters, with 58 data points grouped into clusters and 26 data points identified as noise. The resulting Silhouette Score of 0.376 indicates a cluster structure with a moderate level of separation. The analysis results are further presented through a web-based dashboard to facilitate the monitoring and interpretation of production patterns. This study demonstrates that DBSCAN can assist in identifying production patterns and data points with uncommon characteristics, thereby supporting data-driven decision-making.]]>
Copyrights © 2026
