Predictive Vehicle Maintenance Scheduling And Gis-Integrated Service Recommendation System Using A Machine Learning Approach

Abstract

Vehicle maintenance plays a vital role in ensuring vehicle reliability, operational efficiency, and road safety. Conventional maintenance practices often rely on fixed schedules, manual inspections, or reactive approaches, which may lead to delayed issue detection, unexpected breakdowns, and increased maintenance costs. To address these challenges, this study developed a Predictive Vehicle Maintenance Scheduling and GIS-Integrated Service Recommendation System Using a Machine Learning Approach. The study aimed to design and develop a mobile-based intelligent vehicle maintenance system capable of real-time vehicle monitoring, anomaly detection, GIS-based repair shop recommendations, and automated maintenance notifications. The study employed a project-based developmental research design guided by the Design and Creation Model and utilized the Iterative Prototyping Model for system development. Real-time vehicle diagnostic data, including engine RPM, coolant temperature, battery voltage, fuel level, engine load, vehicle speed, and GPS coordinates, were collected through an OBD-II Bluetooth device connected to an Android mobile application. The gathered diagnostic data were processed and analyzed using the Isolation Forest machine learning algorithm to detect anomalous vehicle conditions and generate predictive maintenance alerts and recommendations. Geographic Information System (GIS) technology was integrated to identify and recommend nearby repair shops, while email and SMS services were utilized to provide automated notifications to users. The developed system was evaluated by fifty (50) private vehicle owners in Misamis Occidental using selected ISO/IEC 25010 software quality characteristics, namely Functional Suitability, Performance Efficiency, Compatibility, Usability, Reliability, and Security. Evaluation results revealed an overall weighted mean of 4.48, interpreted as Very Satisfactory, indicating a high level of user acceptance. The findings indicated that the developed system is functional, reliable, secure, and highly acceptable for predictive vehicle maintenance monitoring and GIS-integrated service recommendations. The study demonstrates the potential of integrating machine learning, OBD-II diagnostics, and GIS technologies to support proactive vehicle maintenance, improve maintenance decision-making, enhance vehicle reliability, and promote road safety.