<![CDATA[Sales of tea powder experience significant fluctuations influenced by consumption trends, seasonal factors, and changes in market demand, which may lead to overstock and understock problems. These conditions highlight the need for an accurate and practical forecasting tool that can directly support managerial decision-making. This study aims to develop a web-based sales and revenue forecasting information system by integrating the Autoregressive Integrated Moving Average (ARIMA) method. The dataset consists of historical tea powder sales from May 2023 to March 2025, aggregated into monthly time series data. The research stages include data preprocessing, visualization, stationarity testing using the Augmented Dickey-Fuller (ADF) test, differencing, ARIMA model identification and parameter estimation, forecasting, and model evaluation using MAE, MSE, and MAPE. The evaluation results indicate that ARIMA(1,1,2) is the best-performing model, achieving an MAE of 4.07, an MSE of 3.15, and a MAPE of 147.34%. The forecasting results show fluctuating patterns in future sales and revenue. The main contribution of this research lies in the integration of the ARIMA forecasting model into a web-based information system that enables automatic and real-time prediction, providing practical support for inventory planning and data-driven managerial decision-making.]]>
Copyrights © 2026
