Mobile broadband performance is an important dimension of digital connectivity because network availability does not necessarily indicate uniform service performance across regions. This study aims to analyze mobile broadband performance, determine the optimal number of clusters, classify Indonesian regencies and cities, and characterize the resulting regional groups based on average download and upload speeds. Secondary data for 2025 were obtained from the Ministry of Communication and Digital Affairs of the Republic of Indonesia. Of 511 regencies and cities in the initial dataset, 345 with complete download and upload measurements were analyzed. The data were standardized using StandardScaler and clustered using K-Means, while the optimal number of clusters was evaluated using the Elbow Method and Silhouette Score. The results identified two clusters, with the highest Silhouette Score of 0.4033 at K=2. The Higher Performance cluster comprised 149 regions (43.19%), with centroid download and upload speeds of 84.272 Mbps and 29.780 Mbps, respectively. The Lower Performance cluster comprised 196 regions (56.81%), with corresponding centroids of 57.233 Mbps and 19.914 Mbps. These findings reveal heterogeneous mobile broadband performance across the analyzed regions and provide a data-driven mapping for evaluating regional disparities in digital connectivity.
Copyrights © 2026