Analyzing Electric Vehicle Charging Behavior for Future Standard and Fast Charging Infrastructure Planning Using Monte Carlo Tree Search

This research investigates the optimization of electric vehicle (EV) charging infrastructure using the Monte Carlo Tree Search (MCTS) algorithm, a powerful computational method designed to address complex decision-making processes under uncertainty. With the rapid increase in EV adoption, the need for strategically located and efficiently managed charging stations both standard and fast is becoming critical. The study harnesses extensive datasets, including historical charging patterns, user behaviour surveys, and real-time sensor data, to model and simulate a range of charging scenarios. Key factors such as charging session duration, state of charge (SOC) levels, and regional variations in EV usage are identified as crucial to understanding and predicting charging behaviour. By incorporating these factors into the MCTS framework, the research develops a predictive model that optimizes the placement and capacity of charging stations. The findings confirm that MCTS is highly effective in forecasting future charging demand and devising strategies for resilient and adaptable infrastructure development. This research contributes significantly to the field by providing a robust framework for the planning and deployment of an EV charging network that is future-proof, supporting the sustainable growth of electric mobility.