SISTEM PERAMALAN PENJUALAN MENGGUNAKAN METODE SINGLE EXPONENTIAL SMOOTHING BERBASIS WEB DI PT. SANGHIANG PERKASA

Abstract

Nutrition Industry, does not have a stock prediction system. So far, PT Sanghiang has relied solely on the warehouse manager's estimates to make decisions about inventory for the next period. This is a problem because stockouts often occur, and sometimes there are insufficient supplies to fulfill orders. Therefore, to avoid such issues, there is a need for a system that can help predict stock levels and make informed decisions in planning future sales strategies. The objective of this research is to develop a web-based sales forecasting system for PT Sanghiang Perkasa and implement the Single Exponential Smoothing method in the forecasting system. The results of this research indicate that the Single Exponential Smoothing method achieves the best forecasting accuracy with an alpha value of 0.2 and a MAPE (Mean Absolute Percentage Error) of 4.01%. The sales forecast for Fitbar Choco Delight 22 G Multigrain in January 2023 is 121,741. Therefore, this method is considered highly accurate as it has the lowest MAPE value.