<![CDATA[Motor Vehicle Tax (PKB) as the primary source of Regional Own Revenue (PAD) in North Sumatra has not been utilized optimally due to the lack of systematic analysis to identify contribution patterns across vehicle categories. This study aims to cluster vehicle categories based on PKB contributions using the K-Means Clustering algorithm, in order to provide clear visual insights for decision-making. The data used were derived from PKB realization across five vehicle categories (Passenger Cars, Buses, Cargo Vehicles, Motorcycles, and Special Vehicles) at 34 SAMSAT Technical Implementation Units (UPT) in North Sumatra for the period January–August 2025. The methodology includes data preprocessing, category aggregation, Min-Max Scaling normalization, and optimal cluster determination using the Elbow method. The results indicate an optimal cluster of K=3 (Silhouette Coefficient=0.62; Davies–Bouldin Index=0.78), with Passenger Cars as the dominant category (71.8%) and Motorcycles at an intermediate level (18.2%). Data visualization through monthly trend charts, proportion heatmaps, and bar charts of the top 10 UPTs reveals seasonal fluctuation patterns in PKB revenue. These findings can serve as a basis for priority collection strategies and data-driven transportation infrastructure planning for regional institutions and SAMSAT North Sumatra]]>
Copyrights © 2026
