<![CDATA[Inventory has a very important role in the company, because it indirectly influences the company's income. If a company does not have inventory, it will experience the risk of not being able to fulfill consumer desires. One data mining technique that can help in processing data to obtain useful information is clustering. The aim of this research is to group inventory of goods, by attributes, initial quantity, quantity sold, and quantity available. Management of inventory data using data mining techniques with the elbow and K-Medoid methods. Then the data that has been grouped can make it easier for stores to determine inventory carefully in terms of procuring stock of goods or products. The results of this research are the use of the elbow method in determining the optimal number of clusters using Python at point 7 (cluster). The clustering results using the k-medoid method with elbow show 7 clusters using the RapidMiner tool. Cluster 0 has 145 products, cluster 1 has 135 products, cluster 2 has 200 products, cluster 3 has 76 products, cluster 4 has 101 products, cluster 5 has 208 products, and cluster 6 has 135 products. Where cluster grouping is based on initial quantity, sold quantity and available quantity with the same or similar value. Clustering results using the k-medoid method without elbows, the clustering process uses 3 clusters with the RapitMiner tool. Cluster 0 has 169 products, cluster 1 has 410 products, and cluster 2 has 421 products. Cluster 0 grouping is based on quantity sold and available quantity, the value is the same, cluster grouping 1 is based on greater quantity sold, and cluster grouping 2 is based on greater quantity available. From the two analysis results it can be seen that the analysis using the k-medoid method with elbows is quite good because in determining the optimal number of clusters using the elbow method and the clustering results in grouping inventory are more effective.]]>
