Author(s): Hajara Sabnam Kareem Navaz

Abstract: CO? emissions from heavy-duty vehicles significantly contribute to environmental pollution, making accurate prediction models essential for sustainable transportation. While various initiatives have been implemented to curb emissions from road transport, the specific impact of heavy vehicles within fleet operations, particularly in supply chain and logistics, is often overlooked. In response to this issue, this paper proposes a machine learning-based prediction model designed to continuously monitor and forecast CO? emissions from heavy-duty vehicles, contributing to a proactive preventive maintenance strategy. The proposed model utilizes the XGBoost Regressor, trained on a comprehensive dataset that includes vehicle-specific parameters and real-time operational data. The results demonstrate the model's effectiveness in real-world applications, achieving a prediction accuracy of 97.88%, along with a Mean Absolute Error (MAE) of 13.29, a Mean Squared Error (MSE) of 422.39, and an R² value of 0.94. These findings underscore the potential of machine learning in enhancing emissions management in supply chain, aiding the development of real-time monitoring systems, and offering a cost-effective approach for fleet operators while contributing to sustainability efforts within the transportation sector.

DOI: 10.33545/26633582.2025.v7.i2b.212

Pages: 160-165 | Views: 407 | Downloads: 193

Download Full Article: Click Here