Artificial Intelligence and Electrical & Electronics Engineering: AIEEE Open Access
Machine Learning-Based Prediction for EV Charging Station Avail-ability and Wait-Time Estimation
Abstract
Anshika Kumari
A simple, effective, and user-friendly charging infrastructure is desperately needed as electric vehicles (EVs) gain popularity. The unpredictable availability of charging stations and possible wait periods provide a major obstacle for EV users, particularly in high-demand metropolitan regions. Current systems often only offer static information about charger locations; they don’t offer real-time availability or forecast usage patterns. In order to forecast EV charging station availability and wait times, this study proposes a machine learning-based prediction model that uses real- time data inputs such station location, charger type, prior usage, traffic conditions, and environmental elements. The proposed approach forecasted station availability with 87.4% accuracy and a root mean squared forecast using Random Forest, Linear Regression, and Long Short-Term Memory (LSTM) models. On average, wait times in crowded cities are 7.8 minutes. These findings show that the approach may reduce wait times and maximise the use of EV infrastructure, offering a reliable way to improve EV user experience and support eco-friendly transportation systems. For lawmakers and urban planners looking to expedite the transition to more ecologically friendly forms of transportation, this study has significant implications.