Peramalan Penjualan Produk Air Minum Dalam Kemasan (AMDK) Kemasan Gelas 220 mL pada Perusahaan XYZ dengan Metode Dekomposisi

This study aims to forecast the sales of bottled drinking water (AMDK) products in 220 ml glass packaging at XYZ Company. The problem faced by the company is the inaccuracy of production planning due to fluctuations in demand, which causes excess and shortage of stock, and has an impact on increasing storage costs and potential delays in distribution. To overcome this, sales forecasting was conducted using historical sales data for 2024 with five time series forecasting methods, namely Moving Average, Simple Exponential Smoothing, Double Exponential Smoothing, Triple Exponential Smoothing, and Decomposition. The evaluation is done using three error measures: Mean Absolute Deviation (MAD), Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE). The analysis showed that the Multiplicative Decomposition method with a seasonal length of 2 is the best method with a MAPE value of 16, MSE 1138, and MAD 2,586,324.