Penerapan K-Means Clustering Untuk Mengelompokkan Tingkat Konsumsi Listrik Menurut Provinsi di Indonesia

Abstract

Electricity consumption in Indonesia continues to increase every year, but its distribution is not evenly distributed across provinces. Therefore, an analysis is needed to group regions based on electricity consumption patterns to help plan more efficient distribution. This study aims to group regions in Indonesia based on their electricity usage patterns, considering customer types such as households, industry, business, social, government buildings, and street lighting. The clustering process is performed using the k-means clustering method. The data used is official PLN data from 2019 to 2024. The analysis process is performed using RapidMiner, with steps of data preprocessing, application of K-Means, and evaluation of the results using the Davies-Bouldin Index (DBI) method. The results of the study show that the provinces in Indonesia are divided into two clusters, namely Cluster 0 with 29 provinces with low-medium consumption and Cluster 1 with 5 provinces with high consumption, especially in the Java Island region. The DBI value of 0.507 indicates that the resulting clustering is quite optimal. These results are in line with the PLN report and should support more targeted infrastructure planning and power distribution policies.