Predicting the Adoption of Clean Energy Vehicles: A Machine Learning-Based Market Analysis
PDF

Keywords

Clean Energy Vehicles
Market Adoption
Machine Learning
Logistic Regression
Random Forest
Decision Tree
EV Market
Sustainability

How to Cite

Hossain, M. ., Rabbi, M. M. K. ., Akter, N. ., Rimi, N. N. ., Amjad, M. H. H. ., Ridoy, M. H. ., Chowdhury, B. R. ., Alam, S. ., & Shovon, M. S. S. . (2025). Predicting the Adoption of Clean Energy Vehicles: A Machine Learning-Based Market Analysis. Journal of Ecohumanism, 4(4), 404–426. https://doi.org/10.62754/joe.v4i4.6742

Abstract

Switching towards clean energy vehicles (CEVs) is a key measure in curbing greenhouse gas emissions and fighting climate change in the USA. Yet, despite mounting environmental consciousness and policy stimulus, the uptake of CEVs is still quite low. The main aim of this research is the creation of a market analysis framework based on machine learning for the prediction of CEV adoption. Utilizing supervised learning algorithms—Random Forest, Logistic Regression, and Decision Tree—the research compares their performance in segmenting prospective CEV adopters in terms of infrastructural, environmental, and socio-economic variables. The dataset included an extensive list of variables designed to capture the various factors that drive clean energy vehicle (CEV) adoption. It includes demographic variables like age, income, educational level, and geographical region, as well as economic variables like vehicle price, purchase incentives, and cost of ownership. In addition, it covers environmental attitudes, captured in terms of questionnaire responses on climate change concerns as well as sustainability values. We initiated this research using a range of machine learning models for the prediction of clean energy vehicle adoption, each of which was used for its particular strengths. To assess the performance of our predictive models, we utilized an extensive range of evaluation metrics: Accuracy, Precision, Recall, F1 Score, and ROC-AUC. Perfect scores on all metrics were recorded for the Decision Tree model, with 100% accuracy, precision, recall, and F1-score. Meanwhile, slightly lower overall performance values were reported for both Logistic Regression and Random Forest models. Sophisticated CEV adoption models' granular outputs can be directly applied in designing and implementing clean vehicle incentive structures at local, state, and federal levels. Knowing the particular socioeconomic, demographic, and geospatial drivers or impediments of adoption in specific regions allows policymakers to craft optimally effective incentive structures. Sophisticated insights derived from patterns of CEV adoption provide irreplaceable value for automotive companies and clean technology firms working in the US market. Future demand for CEVs is important for successful infrastructure planning, especially for the siting of electric charging stations. Monitoring CEV adoption rates is critical for measuring progress towards emissions reduction targets and facilitating broader sustainability planning activities.

https://doi.org/10.62754/joe.v4i4.6742
PDF
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.