Analisis Model Peramalan Permintaan Minuman Kopi Menggunakan Pendekatan Time Series di Coffee Shop

Abstract

Along with the development of the increasingly competitive coffee industry, coffee shops must have a business strategy in managing good coffee supplies to increase competitiveness. Time series forecasting is one of the ways that Koromi Sip & Slurp can use because it serves as a basis for company decision making to develop business strategies. This study aims to determine the best time series forecasting method that can be used as a basis for determining business strategies in managing coffee supplies. The data used is weekly sales data of three types of coffee beans from August 2024 to February 2025. The time series forecasting methods used are Single Moving Average (SMA), Weighted Moving Average (WMA), and Single Exponential Smoothing (SES). Evaluation of the method is carried out based on the forecast error value using the MFE, MAD, MSE, and MAPE methods. The results of the analysis show that the SES method provides the smallest error rate for all types of coffee so it was chosen as the best method. Validation using the Moving Range Chart shows that the forecasting results are within the statistical control limits. Forecasting for the 31st period resulted in a total demand for Sweet Coffee 322 cups, Non-Sweet Coffee 72 cups, and Manual Brew 8 cups. The resulting forecasting is expected to be used as a basis for strategic decision making, one of which is the management of coffee inventory at Koromi Sip & Slurp