Support Vector Regression in Predicting Fuel Efficiency of Electric Vehicles

Addressing the phenomenon of range anxiety among potential EV buyers and current drivers, this paper is organized on a comprehensive exploration of predicting Electric Vehicle (EV) fuel efficiency, guided by a systematic methodology. The initial objective meticulously addressed data identification and pre-processing, providing a robust groundwork for subsequent stages. The successful integration of the Support Vector Regression (SVR) algorithm into the system prototype underscored in delivering accurate predictions, as evidenced explains 73 % of R -squared value shows the potential of SVR in EV fuel efficiency prediction. The creation and evaluation of the SVR-based prototype unveiled strengths in user-friendly design, iterative testing for user alignment and thorough evaluation metrics, emphasizing SVR's adaptability in real-world contexts. However, limitations are noted in the scope of dataset representation and a relatively narrow set of evaluation criteria. Future recommendations focus on enhancing data collection, refining user interfaces, and deploying the system in real-world scenarios. These steps will improve EV fuel efficiency predictions, supporting sustainable mobility, and offering valuable insights and contributions to the realm of sustainable electric transportation.