<![CDATA[The uneven distribution of electrical energy poses a formidable challenge to strategic infrastructure planning and equitable regional development, often hindering sustainable economic growth and exacerbating socio-economic disparities. This issue is particularly acute in a vast and geographically diverse region such as Sukabumi Regency. This study addresses this critical issue by applying the K-Means clustering algorithm to segment 47 sub-districts based on comprehensive electricity customer data from 2019 to 2023, aiming to uncover distinct patterns of energy consumption. The primary novelty of this research lies in (1) its granular application of cluster analysis to sub-district-level electricity customer data for regional energy planning in Indonesia, a previously underexplored area, and (2) the implementation of the results into an intuitive, Streamlit-based interactive web application. This tool serves as a powerful decision-making dashboard for stakeholders, enabling dynamic data exploration and geographical visualization. The methodology encompasses meticulous data collection from official sources, rigorous pre-processing involving data normalization, determining the optimal number of clusters using the well-established Elbow Method, and validating cluster quality with the robust Silhouette Coefficient. The results definitively indicate that three clusters representing Low, Medium, and High energy consumption tiers are the most optimal segmentation. This is substantiated by a strong Silhouette score of 0.6911, which confirms a cohesive and well separated cluster structure. The practical implications are significant, providing a data-driven framework for prioritizing infrastructure investments, enhancing resource allocation efficiency, and supporting the formulation of more targeted energy policies. Ultimately, this study offers a replicable model for other regions facing similar challenges, fostering more sustainable development pathways]]>
