Implementation of Business Intelligence for Sales Prediction and Inventory Optimization and Maintenance in SME Carton Box Company with Open-Source Application

Abstract This study aims to implement business intelligence (BI) as a solution to the challenges faced by the carton box manufacturing company in improving operational efficiency and supporting data-based decision-making. The main problems include inaccuracy in sales forecasting, which causes an imbalance between demand and stock; inefficiency in inventory management, resulting in overstock and stockout; and a reactive maintenance approach, which results in significant machine downtime. In addition to overcoming these problems, this study uses various methods such as Artificial Neural Networks (ANN) to improve the accuracy of sales predictions, Economic Order Quantity (EOQ) and Material Requirements Planning (MRP) for inventory optimization and Overall Equipment Effectiveness (OEE) to improve maintenance efficiency. The Power BI platform is the primary analytical tool to visualize data, identify patterns, and support decision-making. The results of the BI implementation show significant improvements in various aspects of the company’s operations. Sales predictions become more accurate with ANN, which allows the company to manage stock more efficiently and reduce storage costs. Analysis using EOQ and MRP successfully reduced the risk of overstock and stockout, while applying the OEE method increased the machine’s availability, performance, and operational quality. Data visualization with Power BI provides deeper insights for management, facilitates problem identification, and supports faster and more precise strategic decision-making.