CASCAFE Approach With Real-Time Data in Vehicle Maintenance

Service to the vehicle is one of the important factors to be considered as part of vehicle maintenance. The service cost incurred represents how the vehicle has been maintained, making it essential to predict future service costs based on past service records. Vehicle maintenance is crucial for cost management and operational efficiency, as predicting service costs helps optimize resource allocation and long-term planning. Existing methods rely on fixed-interval maintenance schedules or predictive models based on historical usage. However, these techniques often lack adaptability and fail to integrate component aging effectively. Traditional systems are inefficient and costly, leading to inaccurate service predictions, while high variability in repair costs further complicates maintenance planning. This paper introduces CASCAFE (Clustering Analysis for Service Cost and Fuel Efficiency), a clustering-based approach leveraging historical service data for cost estimation. It analyzes kilometers traveled and service cost per kilometer to derive optimal maintenance strategies. Techniques such as Fuzzy C-Means (FCM), Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Gaussian Mixture Models (GMM), and Ordering Points To Identify the Clustering Structure (OPTICS) are utilized to predict future service costs by examining past service data patterns. Among these, Fuzzy C-Means clustering reveals optimal service cost estimation for vehicles with a higher number of kilometers traveled. By integrating component aging and clustering-based analysis, CASCAFE supports informed maintenance decisions and improves cost efficiency in fleet management. Future work will extend CASCAFE to multi-vehicle datasets and incorporate environmental and driver-behavioral factors.