Interpretable Machine Learning for Characterizing Electric Vehicle Charging Behavior: Insights from Real-World Data

As electric vehicle (EV) adoption continues to grow in the United States and worldwide, many have concerns that the grid may not be modernized fast enough to cope with the rapid electricity demand from the transportation sector. This study helps address these concerns by analyzing deidentified real-world driving and charging data from Ford battery electric vehicles (BEVs) collected between 2018 and 2019. These data are used to develop interpretable models that characterize charging behavior and quantify factors most associated with how drivers use charging infrastructure. Prior research has relied on assumptions regarding driver behavior, often overlooking actual charging patterns. By employing generalized linear mixed models (GLMMs), this work offers insights into the factors that influence charging decisions such as next trip distance and state of charge (SOC). The dataset, encompassing over three million park-trip pairs from 1,997 vehicles, reveals that driving behavior significantly influences charging patterns, while infrastructure and regional factors exert a lesser impact. The findings suggest that existing simulation models may oversimplify EV charging behavior assumptions. This research underscores the need for interpretable, data-driven methodologies to inform future EV infrastructure planning and grid management.