Regional Segmentation of Motorcycle Accessory Purchases Using Optimized K-Means Clustering

This study aims to understand and determine differences in consumer behavior when purchasing vehicle accessories from various regions in Indonesia. The K-Means Cluster analysis method was used to group these regions based on recorded purchase data from digital vehicle accessory stores. The data used in this study included a total of 5,097 transactions from October 2023 to October 2024, encompassing various product types, locations, and transaction details. Prior to the analysis, data normalization and categorization were performed to ensure that differences in product names could be identified and the data remained accurate. The analysis revealed distinct regional clusters. The first cluster represents regions with low order volumes and a primary focus on automotive accessories and spare parts. The second cluster demonstrates a high level of interest in components such as tires and vehicle suspension systems. the third cluster, consisting of just two regions, exhibited very high levels of product purchasing activity and a wide variety of products. These findings confirm the diversity of consumer patterns across regions and provide a sound foundation for a data-driven market empowerment strategy for the related industry in Indonesia.