Accurate mass estimation is essential for commercial heavy-duty vehicles (HDVs), as fluctuating payloads significantly impact energy consumption. Precise vehicle mass estimates enhance the accuracy of energy consumption models, leading to more effective energy management systems and performance optimization strategies. For example, improved energy estimates can lead to more optimized routing and refueling schedules, improving operational efficiency and reducing costs. In the context of electrification, accurate mass estimates can inform battery sizing, range predictions, and charging requirements, enabling optimized charge scheduling and seamless integration of electric HDVs into existing operations. While direct mass measurements may be obtained through external weight-in-motion or specialized onboard weighing systems, this paper focuses on methods that use data from Controller Area Network (CAN) systems for alternative real-time predictions. The challenge lies in identifying a method that performs well under the highly variable and often sparse data conditions typical of HDV driving datasets. Three mass estimation approaches are evaluated in this work: k-nearest neighbors (kNN) regressors, road load model-based, and neural networks. Data-driven methods like kNN regressors can perform well with limited data but are computationally intensive during inference. Vehicle model-based methods, grounded in vehicle dynamics, offer high explainability but rely on accurate parameterization and the need for high-resolution data for certain critical calculations. Neural networks, though powerful, require large datasets for effective generalization and lack interpretability. The comparative analysis assesses each method for accuracy, robustness to sparse data, and suitability for real-time application. Telematics data from Class 8 trucks with varying payloads are used for validation, with performance measured through metrics such as mean absolute error (MAE), coefficient of determination (R²), and computational efficiency. Preliminary results indicate that both kNN and neural networks achieve coefficients of determination above 0.99, demonstrating strong predictive capability under diverse conditions.