Abstract—This study proposes a novel privacy-preserving federated learning method, Privacy-Preserving Federated Learning with Federated Averaging (PP-FedAvg), tailored for the oil palm industry to enhance data security while predicting ripeness levels and oil yield of fresh fruit bunches without sharing sensitive data. The primary contribution of this work lies in developing the PP-FedAvg model, which combines encryption techniques with federated learning to ensure secure and accurate classification, achieving 92.5% accuracy across 5209 images of four ripeness levels (raw, semi-raw, semi-ripe, ripe). The prediction results are integrated into a web application, providing practical tools to improve decision-making and operational efficiency in palm oil mills.
 
Keywords—federated learning, privacy-preserving federated learning, oil palm, ripeness, grading, oil palm fruit


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).