This research paper provides an in-depth analysis of the development trajectory of China's New Energy Vehicle (NEV) market, utilizing both the TOPSIS and LSTM models. The TOPSIS model is ingeniously applied to assess the development level of the NEV industry, identifying key factors such as government policies, charging infrastructure, public acceptance, technological progress, and economic indicators. Subsequently, the paper delves into the LSTM model, a specialized form of Recurrent Neural Network (RNN) designed for handling time-series data with long-term dependencies. The LSTM architecture, equipped with forget, input, and output gates, is trained to minimize Mean Square Error (MSE) and Root Mean Square Error (RMSE), thereby optimizing model performance and preventing overfitting. The predictive capabilities of the LSTM model are demonstrated through a forecast of NEV sales in China for the next decade, showcasing the model's effectiveness in capturing market dynamics and growth trends. The integration of TOPSIS for evaluative analysis and LSTM for predictive forecasting offers a robust framework for understanding and anticipating the complexities of the NEV market.