Gendis Jowo experiences fluctuations in the number of nasi box customers, which lead to suboptimal stock management and operational inefficiency, thereby requiring a forecasting approach to predict customer numbers more accurately. This study applies three forecasting methods Single Exponential Smoothing (SES), Double Exponential Smoothing (DES), and Moving Average (MA)—with the aim of determining the most accurate method for forecasting the next period’s customer count. Historical data from January 2022 to August 2025 were analyzed, with SES and DES parameters optimized using the Optimal ARIMA approach, and accuracy evaluated through MAPE, MAD, and MSD. The results show that the Moving Average method with a length of 4 (MA4) provides the highest accuracy with the lowest error values, making it the best-performing model. Based on the MA4 method, the number of customers for the next period is predicted to be 1,065.88, and this result can be used to plan stock requirements, packaging needs, and operational activities more effectively.
Copyrights © 2025