<![CDATA[Inefficient management of tractor spare parts inventory can lead to high storage costs and operational downtime. This issue demands effective calculations to group spare parts based on usage patterns and procurement time. This research aims to apply data mining techniques to cluster tractor spare parts usage using the k-means algorithm to optimize inventory management. The methodology used involves data on spare parts usage over two years, which is then processed using the k-means algorithm to form several clusters based on usage frequency and lead time. This algorithm groups spare parts into clusters that minimize within-cluster variance and maximize between-cluster variance. The formed clusters are interpreted to determine the level of importance of the spare parts and their implications for inventory management strategies. The expected result is the identification of five main clusters grouping spare parts based on usage patterns with very high, medium, and low usage, as well as different lead time variations. These findings are expected to provide important insights for developing more efficient stock management strategies, reducing inventory costs, and increasing the availability of spare parts that match the operational needs of tractors, thus supporting overall efficiency in spare parts usage]]>
