<![CDATA[Palm oil fresh fruit bunch (FFB) production plays a crucial role in plantation management and decision making. However, fluctuations in environmental conditions and plantation characteristics often make yield estimation difficult to perform accurately. This study aims to predict palm oil fresh fruit bunch yield using the Support Vector Machine (SVM) algorithm as a machine learning–based approach. The dataset used in this research consists of monthly production data from 2020 to 2024, including several influential variables such as plant age, land area, rainfall, and soil characteristics. The data were preprocessed through cleaning, transformation, and normalization using the min–max scaling method to ensure consistency and stability during model training. The SVM model was implemented using the Radial Basis Function (RBF) kernel, which is suitable for handling nonlinear data patterns. Model evaluation was conducted by dividing the dataset into training and testing data with a ratio of 80% and 20%, respectively. The performance of the proposed model was measured using Root Mean Square Error (RMSE) and accuracy metrics. Experimental results show that the SVM model achieved an RMSE value of 1.316561 and an accuracy rate of 56.6%, indicating that the model is able to capture the general pattern of palm oil FFB yield data with a relatively small prediction error. Although the accuracy obtained is moderate, the results demonstrate that SVM can be applied as an initial predictive tool for estimating palm oil yield. The findings of this study are expected to support plantation managers in planning harvest activities and optimizing resource allocation.]]>
