<![CDATA[Inventory management in retail businesses has not yet fully optimized the process of goods issuance. The primary problem in managing goods issuance is the risk of loss due to products approaching their expiration dates. This research aims to design and implement a goods issuance recommendation system. The application is developed by implementing the K-Means algorithm to analyze product data. The goods issuance process can be carried out manually or based on recommendations generated through clustering. The recommendation system is integrated into a web-based inventory application. The research methodology encompasses data collection, data preprocessing using aggregation techniques, one-hot encoding for categorical features (type and nature of goods), and feature engineering on expiration dates. The K-Means algorithm is applied to group goods based on similarity. The number of clusters (K) is determined dynamically based on the amount of available data. The system automatically identifies the cluster with the nearest average expiration date as the recommendation target. Clustering results are visualized using Principal Component Analysis (PCA). This system provides end-to-end functionality, ranging from a dashboard and K-Means analysis to the execution of goods issuance. The research results conclude that the system is effective in providing actionable decisions. By prioritizing the issuance of high-risk goods, the system supports operational efficiency and minimizes losses due to expired products.]]>
