The accurate demand forecasting is fundamental to efficient retail supply chain management, helping businesses optimize inventory levels and align operations with consumer needs. This study focuses on forecasting demand for a coffee vending machine using Seasonal Autoregressive Integrated Moving Average (SARIMA) models. To enhance predictive accuracy, weather variables such as temperature, humidity, etc., are incorporated, recognizing their potential influence on coffee consumption patterns. The analysis combines sales data with contextual weather information to develop models capable of capturing seasonal and trend variations more effectively. The forecasting performance is evaluated using Mean Absolute Percentage Error (MAPE) and Mean Absolute Error (MAE) as the primary metrics, with the results showing a MAPE of 32.19% and an MAE of 2.59. These findings demonstrate that the inclusion of weather data significantly improves the model’s reliability. The findings underscore the value of integrating external variables in demand forecasting, offering actionable insights for retailers to refine inventory strategies, minimize waste, and respond to dynamic consumer behavior. By highlighting the interplay between environmental factors and retail demand, this study contributes to advancing data-driven approaches for robust and adaptive forecasting in complex retail environments.