Skripsi
PERANCANGAN DAN EVALUASI KINERJA MESIN PEMISAH KEMATANGAN BUAH KOPI DENGAN SENSOR TCS3200 DAN MIKROKONTROLER ARDUINO
The racutan harvesting method and manual sorting of coffee cherries often result in low quality and inefficiency in terms of time and labor. This occurs because unripe cherries are mixed with ripe ones, thereby reducing export quality, while manual color-based sorting is highly subjective. This study aims to design and evaluate the performance of an automated coffee cherries ripeness sorting machine based on an Arduino microcontroller and a TCS3200 sensor, integrated with a conveyor system to measure sorting capacity and test the accuracy of ripeness detection. This research employed an experimental method, with data presented descriptively. The performance of the sorting machine was evaluated based on several parameters, including the accuracy of RGB color readings by the TCS3200 sensor, the sorting time per cherry, and machine accuracy calculated using the MAE and MAPE methods to ensure consistent quality in the sorting process. The results showed that the machine achieved an average accuracy of 92.33%, with an TIR value of 96%, an MAE value of 0.08, an MAPE value of 8.3%, and an average sorting time of approximately 5.2 seconds per cherry. Therefore, this machine is useful for automating the coffee cherry sorting process, reducing time and labor requirements, and supporting modern and sustainable agricultural processing systems.
No other version available