OPTIMIZATION OF INVENTORY MANAGEMENT USING THE FUZZY C-MEANS ALGORITHM

Abstract

ABSTRACTS

To analyze inventory data, this study utilizes the Fuzzy C-Means clustering algorithm. The Fuzzy C-Means clustering results in ten groups, each labeled from Cluster 0 to Cluster 9, with varying numbers of items in each cluster. The largest clusters are Cluster 3 with 36 items and Cluster 4 with 34 items. Further examination of each cluster reveals consistent patterns and characteristics, aiding in understanding the differences among items within each group. According to the grid parameter evaluation, the number of clusters (K=10) provides the best results with a low Davies Bouldin index, indicating good clustering quality. The scatter plot shows how items are distributed across each cluster. Clusters 3, 6, and 7 stand out as the groups with the most items. Ten clusters with patterns that may not be immediately apparent are created using the Fuzzy C-Means method. Clusters 3, 6, and 7 continue to be the groups with a significant number of items. An in-depth analysis of each cluster provides specific insights into stock behavior and potential business opportunities. To achieve Objective 2, the clustering analysis results help Nata Frozen Food optimize their business strategies. Marketing strategies can be adjusted, inventory management can be improved, and product portfolio development can be enhanced. The evaluation results indicate that the appropriate number of clusters can improve operational efficiency and supply chain effectiveness by balancing model complexity and clustering quality.