Low-Cost Predictive Maintenance Modeling for SMB Fleets Using Operational Data

This research explores the application of low-cost predictive maintenance (PdM) models for small and medium-sized business (SMB) fleets, leveraging readily available operational data. SMB fleets often lack the resources for sophisticated PdM systems. This study investigates the feasibility of using easily accessible telematics data, such as mileage, fuel consumption, and basic engine diagnostics, to predict component failures and optimize maintenance schedules. We compare the performance of several machine learning algorithms, including logistic regression, support vector machines (SVM), and random forests, in predicting failures of critical fleet components. The models are trained and validated using a real-world dataset from a diverse SMB fleet. The results demonstrate that even with limited data and computational resources, effective PdM models can be developed to reduce downtime, lower maintenance costs, and improve the overall operational efficiency of SMB fleets. Furthermore, the study provides a framework for SMBs to implement these models using open-source tools and cloud-based platforms, thus minimizing upfront investment. The implications of this research are significant for SMBs looking to enhance their fleet management strategies through data-driven decision-making.