Comparison of Single Exponential Smoothing Method With Double Exponential Smoothing Method Prediction of Salt Sales

Abstract

Predicting the quantity of product sales in the future aims to control the amount of existing product stock, so that the shortage or excess of product stock can be minimized. When the quantity of sales can be predicted accurately, the fulfillment of consumer demand can be managed in a timely manner and the company's cooperation with consumers is maintained properly so that the company can avoid losing sales and consumers. This study aims to analyze the accuracy of predicting the quantity of sales of salt using the Single Exponential Smoothing (SES) method compared to using the Double Exponential Smoothing (DES) method, so that a more accurate method will be obtained for predicting the quantity of sales. The results of testing the comparison of the level of accuracy can be done by evaluating the error value of the forecasting results with the Mean Absolute Percentage Error (MAPE). The lowest MAPE result obtained is in the SES method when the parameter α = 0.054 with a MAPE result of 7.932% which means the accuracy value is very accurate. Whereas with the DES method the MAPE value is 28.145% while the parameter α = 0.845 β = 0.214 which means the value of accuracy is reasonable. Based on the MAPE results obtained using the two methods above, the Single Exponential Smoothing method is more accurate for use in predicting salt sales. Whereas with the DES method the MAPE value is 28.145% while the parameter α = 0.845 β = 0.214 which means the value of accuracy is reasonable. Based on the MAPE results obtained using the two methods above, the Single Exponential Smoothing method is more accurate for use in predicting salt sales. Whereas with the DES method the MAPE value is 28.145% while the parameter α = 0.845 β = 0.214 which means the value of accuracy is reasonable. Based on the MAPE results obtained using the two methods above, the Single Exponential Smoothing method is more accurate for use in predicting salt sales