Analisis Peramalan Produksi Kelapa Sawit Pada Perkebunan Rakyat di Provinsi Kalimantan Barat

Abstract

Oil palm is one of the largest agricultural products in Indonesia and has high economic value and can improve the welfare of oil palm farmers. The amount of oil palm fruit production is not always stable or increasing, instead it fluctuates and falls which is influenced by many factors. This study aims to estimate the production level of palm fruit for the next five years based on data for 2012 - 2021 in West Kalimantan Province and to compare forecasting the amount of palm oil production using 3 (three) time series forecasting methods, namely the linear trend method, single exponential smoothing and moving averages. The image processed in this study is the production of oil palm fruit in the last 10 years which is sourced from data from the Central Bureau of Statistics based on the West Kalimantan Provincial Plantation and Livestock Service. Forecasting accuracy is then measured using MAD (Mean Absolute Deviation), MSE (Mean Squared Error) and MAPE (Mean Absolute Percent Error). The projection of the average amount of palm oil production in the next 5 years (the next period) using 3 forecasting methods is: 16,435,545 (linear trend); 10,994,812 (single exponential smoothing) and 12,487,150 (moving average). Accuracy measurements using MAD, MSE and MAPE show that the most accurate method for forecasting the amount of palm oil production is forecasting using the linear trend polynomial method (MAD 108,855.10; MSE 19,896,565,928.90; and MAPE 1.29) because it has a high error rate. smaller than the results of forecasting using the moving average method and single exponential smoothing.