In order to discover the typical process route (TPR) in the Computer Aided Process Planning (CAPP) database, Hierarchical Clustering algorithm is adopted. A mathematics model based on the data matrix was built to describe the process route (PR). On the base of the operation code, the distance between operations, between the PRs, between clusters were measured to evaluate the PR similarity. Then, the PR clusters were eventually merged by the hierarchical clustering algorithm. Three methods are listed to confirm the clustering granularity that determines the clustering result. This TPR discovery method is successfully applied to discover the axle sleeves’ TPR.