Unveiling the Green Revolution: Predicting CO2 Emission and Fuel Economy of Vehicles for a Sustainable Future

Abstract In this dynamic and ever-evolving world, the need for sustainable transportation solutions has become increasingly important. This research employs a data-driven approach aimed at the accurate prediction of two pivotal factors in vehicle sustainability: CO2 emissions and fuel economy. Through comprehensive exploratory data analysis, this study unveils significant findings that underscore the influence of vehicle characteristics and fuel types within its scope and tackling the complex problem of modeling environmental efficiency with robust algorithms. In this work, machine learning models were developed, employing a diverse set of algorithms including the Random Forest, Decision Trees, K-Nearest Neighbor and Artificial Neural Network. The models were rigorously evaluated using metrics such as the R2 score and Root Mean Squared Error. Notably, the implementation of cross-validation was integral in assessing model generalization and mitigate overfitting. This research findings show that the KNN and Random Forest models emerged as the most stable and consistent across CO2 predictions. The Decision Tree and Random Forest models emerged as the most stable and consistent across fuel cost predictions. These models have demonstrated high R2 scores, closely matching the actual values and generalizing well across the 10-fold cross validation. Adopting these models could significantly improve environmental impact assessments and support the creation of eco-friendly vehicles. This study advances sustainable transportation by introducing a machine learning approach for improving vehicle emissions and fuel efficiency predictions. The analysis and practical application of the best model offers engineers a novel tool for designing more sustainable vehicles and contributes to the decision-making process regarding environmental policies in the automotive sector.