Electric vehicle (EV) transition is the key to sustainable transport in greenhouse gas emissions. In this paper we use historical EV sales (20152023), demographic, policy and infrastructure, and fuel prices to project market growth to 2030. We fit and contrast four models of forecasting, such as Linear Regression, XGBoost, ARIMA, and Facebook Prophet, and use SHAP (SHapley Additive exPlanations) to explain the results of the model. According to our findings, XGBoost demonstrated the best performance (MAE = 3.92, RMSE = 5.47, R2 = 0.89), where charging-station density and government incentives proved to be the most significant forces that promote adoption. The novelty of the work is in the combined application of SHAP-based interpretability with a multi-model benchmarking scheme, which gives practical, data-grounded advice to policymakers and stakeholders within the industry.