Main Menu Sales Forecasting at Bakso Pak Eko MSME Using the Single Exponential Smoothing Method

Abstract

Sales volume is a critical factor influencing business sustainability, particularly for Micro, Small, and Medium Enterprises (MSMEs) operating in the perishable culinary sector. Bakso Pak Eko MSME faces persistent operational challenges in daily production planning, which currently relies on manual intuition and informal estimation, frequently resulting in inventory surpluses, ingredient spoilage, or unmet consumer demand. To address these operational inefficiencies, this study develops a responsive web-based sales forecasting system for primary menu items using the Single Exponential Smoothing (SES) method to support data-driven production scheduling. The SES method was selected due to its mathematical simplicity, low computational overhead, and proven efficacy in processing short-term historical sales records that exhibit stationary demand patterns without pronounced trend or seasonal fluctuations. The application was engineered using Python and the Streamlit framework, integrating dynamic menu catalog administration, daily transaction logging, customizable smoothing constants, automated forecasting routines, and real-time gross revenue estimation. Functional reliability was verified through Black Box Testing, while predictive performance was benchmarked using the Mean Absolute Percentage Error (MAPE) metric across three core menu items: regular meatballs, egg meatballs, and jumbo meatballs. The evaluation results demonstrate that the application operates reliably across all functional workflows without runtime anomalies. Configured with an optimal smoothing constant of α = 0.3, the model achieved a MAPE value of 1.56%, categorizing the forecast as highly accurate with an absolute deviation of only one to two portions per menu item. Furthermore, directly translating unit portion projections into daily revenue calculations provides the enterprise owner with actionable financial transparency alongside physical preparation targets. Consequently, this system effectively bridges computational modeling and shop-floor decision-making, enabling MSME managers to streamline raw material procurement, mitigate food waste, and enhance overall operational cost-efficiency.