Content area
Full text
Sales of electric (EV) and hybrid electric (HV) vehicles in the United States increased by approximately 83 percent and 76 percent respectively in the past year.1 The U.S. government will invest approximately $7.5 billion USD to expand access to EV chargers with a long-term goal of establishing a national charging network.2, 3 With this investment influx in the EV domain, it is important to incorporate available data sources to ensure strategic and equitable investments in EV chargers and monitor evolving usage patterns for agile adaption of public policy as the charging network is built out.
Data Driven Philosophy
The paper uses an August 2021 data set of connected vehicles (CV) for the U.S. states of California, Connecticut, Georgia, Indiana, Minnesota, North Carolina, Ohio, Pennsylvania, Texas, Utah, and Wisconsin to analyze EV and HV usage at the state, city, and county level. In this research, a CV refers to a vehicle that shares select on-board data with the original equipment manufacturer (OEM) in real-time. This research focused on identifying fast-charging station gaps on long range travel corridors, measuring dwell times near public charging stations, and developing comparative metrics that can be used for inter- and intra-state comparisons. It is believed that the tables, figures, and methodologies in this paper will stimulate further dialogue among decision makers and transportation professionals to build upon these CV data-driven methodologies to inform public policy and investments.
Study Data and Summary Statistics
Table 1 summarizes the coverage of the CV data set utilized as well as details regarding vehicle miles traveled (VMT) by EV and HV vehicles in those 11 states. Although this is not an exhaustive or unbiased data set, it is a nationally available CV data set that can be obtained in near real-time. Cloud storage and data warehousing services were used to manage and analyze this big data set with modest costs and in a manner that can be easily scaled to all 50 states.4
A single connected vehicle trajectory is obtained as a series of waypoints at 1 to 3-second fidelity with a 3-meter (9.8 feet) geolocation accuracy. Each waypoint provides a timestamp, speed, and heading value for the vehicle in addition to a vehicle classification code that helps identify the make and model by using the...