<![CDATA[This research utilizes the K-Means clustering algorithm to analyze car rental data from PT. Station Armada Indonesia, aiming to simplify customer car selection and improve the company's market responsiveness. The study addresses the problem of customer confusion stemming from the wide variety of car types offered by the company. By employing K-Means clustering on August 2023 rental data, the research groups cars based on rental price and mileage. The dataset, initially encompassing four car categories (Minibus MVP, SUV, City Car, and Van/Bus), was further detailed to include individual car models. Three parameters—rental duration, rental price, and mileage—were used for clustering. The K-Means algorithm, chosen for its ease of implementation and speed, was applied iteratively using Euclidean distance to assign data points to the nearest centroid. The study initially defined two clusters. Manual calculations, detailed in the paper, demonstrate the clustering process. These manual results were then compared against results obtained using RapidMiner Studio version 10.1, showcasing the software's efficiency in handling the K-Means process. The RapidMiner output included Data, Statistics, and Annotations views, providing a comprehensive analysis of the clusters. The final clustering, achieved after three iterations, revealed two distinct clusters: one representing less popular car types (Cluster 0), and the other representing the most popular car types (Cluster 1). Cluster 0 contained six car types with average customer mileage ranging from 673 km to 2050 km, while Cluster 1 included 24 car types with average mileage between 270 km and 3388 km. The findings enable PT. Station Armada Indonesia to optimize fleet management and marketing strategies by focusing on the most in-demand car types. The study concludes that K-Means clustering, implemented via RapidMiner, offers a valuable tool for enhancing customer understanding of car selection and improving the company's overall efficiency.]]>
