Double Exponential Smoothing Method to Forecast the Production of Tembakau in North Sumatra

Abstract

Forecasting is the process of using mathematical calculations and quantitative data from the past to predict situations that have not occurred or will occur in the future. Determine the strategy used in pursuing choices based on several things that have been done so that it is very feasible for planning. Brown's Double Exponential Smoothing, especially the smoothing approach with alpha values of 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9 will be used as a forecasting technique in this study. Using tobacco production data from 2012 to 2021, the forecasting results show that the best guess value achieves the smallest Mean Absolute Precentage Error (MAPE) of the nine alpha values used. with the parameter alpha = 0.2 produces the lowest error rate of 21.37 percent. The form of the forecasting equation is Fi+m = 1666.84892857549 + 86.604473843712 m. That is, the accuracy of forecasting the amount of tobacco production in the next ten years is 65.37 percent.