Utilization of Big Data in Oil Palm Plantation to Predict Production Using Artificial Neural Network Model

Abstract

The oil palm plantations in Indonesia are more than 14 million hectares and have been cultivated for more than 100 years in various types of land, climates, and various technical cultural treatments. The cultivation process will produce very large data. However, the utilization of these data has not been optimal and is still being managed partially. In the 4.0 industrial revolution, big data is a key asset in building artificial intelligence to support precision agriculture. One of the uses of big data is to build predictive models. An artificial Neural Network (ANN) is a model that can be used to predict by utilizing big data. On the other hand, production prediction is a very important activity to help planters in making decisions on all plantation activities. This study aims to use big data in oil palm plantations to predict production using ANN. The input data used in this study are components that have an influence on production. Meanwhile, the output to be predicted is annual yield and FFB production. The ANN model used is multilayer perceptron backpropagation with architecture 24-25-35-25-1. This model can accurately predict annual yield and total production based on block, division, estate, palm age, and progeny with MAPE and R are 10.52 % and 0.96 respectively.