<p>The rising adoption of electric vehicles (EVs) minimizes greenhouse gas emissions but exerts pressure on power grid infrastructure. Assessing the requirements for electric vehicle charging can help mitigate this issue. Implementing a price strategy may redirect the demand for charging services from off-peak hours to peak hours, thereby enhancing the profitability of charging stations. The study aims to present a strategy utilizing a machine learning algorithm to forecast the demand for charging stations and create a model for determining electric tariff pricing. This research analyzes the ACN dataset, one of the limited publicly available datasets for non-residential electric vehicle charging. Additionally, several machine learning techniques, such as Random Forest, XGBoost, LSTM, and ANN, were utilized to forecast the demand for charging stations. Furthermore, the ANN technique outperformed other machine learning methods by adeptly discerning intricate correlations and facilitating precise predictions. The proposed machine learning approach establishes a robust basis for forecasting the demand for charging stations utilizing past data. This model assists industry and policymaker stakeholders in formulating strategies for forecasting electricity consumption and developing electric vehicle infrastructure. It supports the electric vehicle supply chain and fosters a more sustainable future by enabling the coordination and promotion of electric vehicles and the widespread use of electric transportation.</p>