Plug In For Lunch: Exploring Dining Options Around American EV Charging Stations

Seung Gyu Baik, Ahyoung Chang, Junfeng Jiao

Keywords: LLM, EVCS, Urban Planning

TL;DR: An analysis with AI and GIS on EV charging stations in the U.S. reveals that those stations can be strategically deployed around underserved cuisine types, enhancing driver-business interaction and generate economic impacts in local communities.

Abstract: Drivers of Electric Vehicles (EVs) must spend some time around EV Charging Stations (EVCS). This on-site dwelling time can create opportunities for EV drivers to interact with urban amenities and nearby businesses. This driver-business interaction is known to have a positive spillover effect on the local economy by attracting high-income drivers to low-income areas. Using methods in Geographic Information Systems (GIS) and Artificial Intelligence (AI), this paper reviews the status quo of EVCS placement regarding this interaction, across the Contiguous United States. We use dining locations, such as restaurants and bakeries, as a mediator of this interaction. The shortage of dining options is apparent in regions where businesses are sparsely and linearly spread out along the road network without a dense commercial core. A Large Language Model (LLM)-assisted analysis of the type of cuisine each dining location serves reveals room for improvement in the access to certain cuisine types. Based on these observations, this paper proposes an initiative to strategically deploy EVCS around dining locations that serve low-access cuisine types. This can contribute to destination development for EV drivers and encourage driver-business interaction for local business owners, potentially delivering an economic boost to the surrounding community. This paper demonstrates the application of AI and LLM to loosely-formatted, crowdsourced geospatial data and the practice of urban planning.

Submission Number: 64