Sales Forecasting of Betung Bamboo Shoots (Dendrocalamus asper) by the use Triple Exponential Smoothing

Abstract

Betung bamboo shoots (Dendrocalamus asper) is one of the processed fresh products sold at Tiara Dewata Supermarket, where the processed products are divided into three, namely raw bamboo shoots, chopped bamboo shoots, and ordinary bamboo shoots. The shelf life of bamboo shoots is very short, only 1-3 days. Furthermore, the monthly sales for these three fresh products fluctuate and it is difficult to predict the trend. Therefore, a forecasting method is needed in order to minimize the losses that will occur. The purpose of this study was to find the best alpha value that can be used to obtain the best time series forecasting data for the next one year for the three types of Betung bamboo shoots using the TriplemExponential Smoothing method. The data used in this study is the actual sales data of the three processed bamboo bamboo shoots from March 2019 - May 2020. The best alpha value that can be used for forecasting is the calculation of time series data with an alpha value of 0.1 - 0.9 which has a value the smallest error, where alpha 0.4 in raw shoots with an error value of MSE 10.308, RSME 3.287, MAPE 19%, alpha 0.4 in chopped bamboo shoots with an error value of MSE 113.809, RMSE 10.688, MAPE 5%, and alpha 0,4 on ordinary shoots with an error value of MSE 511,902, RMSE 22,625, MAPE 4%. From the calculation using the alpha value, it can be concluded that the triple exponential smoothing method with alpha 0.4 is used to predict the sales time series data of the three processed Betung bamboo shoots from the period June 2020 - May 2021.