Anggi Vasella, - (2023) Penerapan Triple Exponential Smoothing dan Arima dalam Memprediksi Produksi Crude Palm Oil. PENERAPAN TRIPLE EXPONENTIAL SMOOTHING DAN ARIMA DALAM MEMPREDIKSI PRODUKSI CRUDE PALM OIL, 8 (2). pp. 236-246. ISSN 2541-1004

Preview

Text (JURNAL) ANGGI VASELLA.pdf - Published Version Download (1MB) | Preview

Abstract

Abstract In the field of oil palm plantations, PT.XYZ, located in Riau Province, is a company that produces one of them Crude Palm Oil (CPO). It is known that from the last 10 years of production, the selling price tends to be unstable resulting in stockpiling. Therefore, forecasting the right amount of production is carried out so that the problem of stockpiling can be overcome, so this study aims to forecast CPO production stock using a comparison of two algorithms, namely Triple Exponential Smoothing and ARIMA. Forecasting involves taking historical data and predicting it for the next period. After the forecasting process is carried out, the level of error in forecasting is tested using the Mean Absolute Percatage Error (MAPE) method to show the range of error values in forecasting calculations based on the smallest error. The output after testing with the TES method gets a benchmark αlpha = 0.5, βeta = 0.004, and gamma γ =1.0 the error rate is obtained using MAPE accuracy of 10.1% and 1.4% for the ARIMA model. The output of the TES method gets the MAPE category with good forecasting ability and while the output of the ARIMA method is included in the MAPE category with very good forecasting ability according to the MAPE range assessment. The role of this research is needed in order to provide information to related companies as an additional reference in forecasting CPO production. The results of the research conducted obtained the ARIMA method is the best method with the smallest error calculation of the MAPE value.

Actions (login required)

View Item