Forecasting and Analyzing Vehicle’s Energy Resources Utilization for Improved Performance

This project centers on forecasting vehicle fuel usage to improve efficiency and realize economic and environmental gains. It provides an in-depth analysis of data-driven approaches, focusing on critical elements like vehicle dynamics, driver behavior, and environmental conditions. Traditional machine learning models such as SVM and random forests are compared with neural network-based models, including artificial and deep computational frameworks, which excel in handling large datasets, training efficiency and ensuring prediction consistency. Here, the project develops a Logistic model, utilizing features such as Mass Air Flow, Vehicle Speed, Revolutions Per Minute. The proposed model provides a direct but effective solution, offering dependable results and serving as an alternative to more complex algorithms. This study highlights the potential of such models in optimizing fuel consumption and enhancing vehicle performance, laying the groundwork for future innovations in predictive modeling.