Risk prediction in EV Batteries Supply Chain using Machine Learning Approaches

The escalating concern for the environment has led to a heightened demand for electric vehicles, driven by the imperative to establish a more environmentally sustainable supply chain. While the utilization of electric cars offers a commendable reduction in air pollution and irreversible fuel consumption, the entire lifecycle of lithium batteries, encompassing raw material extraction to final battery recycling, engenders environmental pollution and societal disenfranchisement, particularly in specific regions where material properties and legal frameworks diverge. This study endeavors to employ machine learning methodologies to anticipate Environmental, Social, and Governance (ESG) scores, considering diverse factors within the context of lithium batteries. The findings indicate that human rights abuse emerges as the most influential factor influencing the ESG score. Moreover, advanced machine learning techniques such as Support Vector Machines (SVM) and Neural Networks demonstrate heightened efficacy in accurately classifying data, particularly in scenarios with a limited number of imbalanced generated datasets.