<![CDATA[Indonesian palm oil cooperatives commonly rely on manual recording, resulting in fragmented data and limited product traceability. This study proposes an Integrated Smart Traceability Framework that combines digital information systems with an AIbased anomaly detection module. Design Science Research (DSR) was adopted as the methodology, the PIECES Framework guided requirement elicitation, and Isolation Forest was used to identify anomalous harvest transactions. The threelayer architectureDigital Information System, Artificial Intelligence, and Traceability Enginewas implemented as a Laravelbased web application following the MVC pattern. Black Box Testing confirmed 100% functional success across 14 test scenarios covering 10 functional modules. On 500 simulated harvest records, the AI component achieved 94.0% accuracy, 64.3% precision, 90.0% recall, 75.0% F1-score and 0.967 AUC-ROC. The framework provides a traceability mechanism that supports preparedness for compliance with the European Union Deforestation Regulation (EUDR). However, this evaluation was conducted on simulated data and a single cooperative; the AI performance results have not yet been confirmed with real operational data and further testing is required.]]>
Copyrights © 2026
