Implementation of the Single Moving Average Method for Web-Based Basic Goods Stock Prediction in Nginjen Makmur Supermarket Business

Inventory management is an important aspect of retail business operations, as inaccurate stock planning can lead to overstocking or shortages. This study aims to develop a web-based staple goods stock prediction system for Nginjen Makmur Supermarket using the Single Moving Average (SMA) method. The research uses 17 months of historical sales data covering four commodities: rice, sugar, cooking oil, and curly red chili. The system was developed using PHP and MySQL, with features for managing product and sales data, generating stock predictions, and presenting sales and prediction reports. Forecast accuracy was evaluated using Mean Absolute Deviation (MAD) and Mean Absolute Percentage Error (MAPE), with SMA calculation windows of three and six periods. The results reported an aggregate MAPE of 2.68% for the three-period window and 2.38% for the six-period window. Both values were below 10%, indicating very high accuracy based on the evaluation criteria used in the study. The developed system is expected to support more structured and data-driven inventory planning at Nginjen Makmur Supermarket.