Application of Oil Palm Dissection Method to Predict Bunch Production in Commercial-Scale Oil Palm Plantation

Abstract

Prediction of the annual oil palm yield is an important decision-making tool in harvesting operational management, management of Palm Oil Mill (POM), trading of Crude Palm Oil (CPO) and its derivative products along with management of POM waste. The palm census method has a better accuracy rate than using previous year's production data. Calculation the bunch number on the tree is used as an annual prediction by using the Oil Palm Dissection (OPD) method. OPD is carried out to calculate the number of inflorescence and bunches as a basis for predicting bunches for the next 18 months. The purpose of this study is to analyze the application of the commercial scale OPD in bunch prediction activities in oil palm plantations. This research was conducted on oil palm aged 13 -15 years (plateau yield phase). The research stages include sample selection, tree dissection, inflorescence and bunch modeling, as well as testing the accuracy of the OPD compare to actual production. Prediction-accuracy test uses Mean Absolute Percentage Error (MAPE). The accuracy of OPD in predicting the bunch number is 70% (feasible criteria) and the prediction of oil palm yields production is 81% (good criteria) in oil palm plantations on a commercial scale. Prediction of OPD needs to consider some correction factors such as climate, environment, Average Bunch Weight (ABW), palm age, and soil type, to improve the validation result and accuracy of predictions. The percentage of error in the bunch number between 10 -32% of the OPD prediction can be combined with other factors to determine the value of the correction factor for bunch loss and oil palm yield production.