Application of an inductive sensor system for identifying ripeness and forecasting harvest time of oil palm

Ripeness estimation of oil palm fresh fruit bunches is a crucial component in the management of oil palm harvesting, as it will lead to profitability and marketability of the product. The purpose of this study is to develop an oil palm maturity detection device that is not only able to identify oil palm maturity, but also predict harvest time. In this work, the resonant frequency data were collected using an inductive sensor from a total of 600 fruits at various ages of ripeness. Intelligent algorithms are embedded into the system to recognize the oil palm ripeness, i.e., discriminant analysis for identifying ripeness and polynomial regression for forecasting harvest time. In oil palm plantations, we prepared 55 fresh fruit bunches to identify their ripeness and forecast harvest time. Based on the field test performance, the inductive sensor system can determine the oil palm ripeness with an accuracy of 100 % and forecast the harvest time with RMSE of 13.45. Therefore, the proposed system has the potential to be implemented in the evaluation of harvesting oil palm due to its various advantages, i.e., being accurate, rapid and non-invasive.