Abstract:
Electric vehicle (EV) users living in colder or warmer climates experience shorter traveling ranges, slower acceleration, and longer recharge times, which might discourage the adoption of EVs. Using California and New York data, we conducted a study in the United States (US) to explore the relationship between temperature and EV adoption. We collect land surface and air temperature data at the ZIP code level in addition to sociodemographic, charging infrastructure, and land cover data. We then use random forest machine learning to predict battery electric (BEV) and plug-in hybrid electric (PHEV) vehicle population change rate and penetration. Our findings reveal that temperature is the most important predictor of BEV and PHEV population change rate and penetration. Specifically, average daily mean temperature variation emerged as the most influential factor in the variable importance analysis for BEV and PHEV adoption, with December and January average temperatures also ranking among the top five factors. Understanding the interplay between temperatures and EV adoption is crucial to assess the feasibility of adopting EVs in different climatic regions and ensure equitable access to this technology. This study highlights the importance of considering environmental factors in designing geotargeted interventions to promote EV adoption and sustainable transportation options.
See publication:
https://ieeexplore.ieee.org/abstract/document/10485540This publication pertains to:
Systems of SystemsPublication Authors:
- Gaia Cervini
- Jinha Jung
- Konstantina Gkritza