<![CDATA[Driver performance assessment is a critical aspect of transportation company operations; however, manual evaluation tends to be inefficient and prone to subjectivity. This study aims to classify driver performance at a transportation company in Surabaya using the K-Means Clustering algorithm based on historical delivery data from 2020 to 2025. The dataset comprises 31 drivers with 135 data records. The research workflow includes data collection, preprocessing through aggregation and Min-Max Scaling normalization, optimal cluster determination using the Elbow Method, clustering process, and evaluation using Silhouette Score and Davies-Bouldin Index. Results indicate that k=3 is the optimal number of clusters, categorizing drivers into three performance groups: High (7 drivers, 22.6%), Moderate (15 drivers, 48.4%), and Low (9 drivers, 29.0%). A Silhouette Score of 0.5948 and Davies-Bouldin Index of 0.5191 confirm that the clustering results are valid and of good quality. These findings are expected to serve as an objective, data-driven foundation for management in driver performance evaluation and development.]]>
Copyrights © 2026
