<![CDATA[Background: The Class II Non-TPI Muara Enim Immigration Office processes thousands of passport applications annually. However, existing data utilization remains limited to administrative recording and routine reporting, constraining its strategic value for service planning. Objective: This study aims to identify meaningful patterns in passport application data and determine the optimal cluster structure using the K-Means algorithm to support evidence-based decision-making. Methods: Therefore, this study was conducted to apply data mining techniques using the K-Means clustering method to analyze passport application patterns based on data from 2022 to 2024, with attributes such as gender, travel purpose, type of application, type of passport, and year of submission, implemented through Python with the support of Pandas, NumPy, Matplotlib, Seaborn, and Scikit-learn libraries. Results: The results show that passport applicants can be grouped into several clusters with specific characteristics — for instance, clusters dominated by applicants of productive age with purposes of working or performing Umrah, as well as other clusters consisting mainly of applicants traveling for tourism or education; moreover, the analysis reveals spikes in applications during certain periods, such as before the Hajj season and long holidays. Conclusion: Thus, it can be concluded that applying K-Means clustering provides added value beyond conventional data processing, enabling administrative data to be transformed into informative and predictive patterns that support resource planning, staff allocation, and more effective data-driven service policies.]]>
