ArticlePublisher preview available

Built environment and public electric vehicle charging: an investigation using POI data and computer vision

Authors:
Junfeng Jiao at University of Texas at Austin
Junfeng Jiao
Seung Jun Choi at University of Texas at Austin
Seung Jun Choi
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

Public electric vehicle charging stations (EVCSs) are vital for boosting EV adoption. This study investigates Seoul’s public EV charging patterns, taking into account the surrounding urban built environment. We collected built-environment data from land-use maps, Point of Interest (POI) data, and panorama images near public EVCS. The computer-vision technique was used to extract scene features from panorama images. We conducted a spatiotemporal analysis of public EVCS usage. The built-environment factors underwent dimensionality reduction and were assessed for outliers. Descriptive analysis revealed afternoon peak charging times and variations between chargers. Additional peaks are observed in the weekday late evening for chargers located near mega-retail stores. Public EVCS in Seoul were utilized more on weekdays than on weekends. Public EVCS in central business districts saw the most significant usage, with potential cases of overuse. An analysis of the built environment around the chargers showed unique characteristics, with some forming identifiable clusters. The most used public EVCS had more parking areas than other POIs, matching visual observations. Computer visioning mainly recognized highways, parking lots, and crosswalks as common features near the chargers. Outlier test results generally defined fast chargers in the central business district area as outliers. The results also demonstrated that built-environment measures from POI data and computer vision can be used in a complementary manner. Our study offers empirical findings to enhance the understanding of public EV charging usage. We demonstrated the use of POI data and computer-vision techniques to quantify the built environment.
Figures - available from: Public Transport
This content is subject to copyright. Terms and conditions apply.
A preview of this full-text is provided by Springer Nature.
Learn more
Content available from Public Transport
This content is subject to copyright. Terms and conditions apply.
Vol.:(0123456789)
Public Transport
https://doi.org/10.1007/s12469-024-00383-6
CASE STUDY ANDAPPLICATION
Built environment andpublic electric vehicle charging:
aninvestigation using POI data andcomputer vision
JunfengJiao1 · SeungJunChoi1
Accepted: 28 November 2024
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025
Abstract
Public electric vehicle charging stations (EVCSs) are vital for boosting EV adop-
tion. This study investigates Seoul’s public EV charging patterns, taking into
account the surrounding urban built environment. We collected built-environment
data from land-use maps, Point of Interest (POI) data, and panorama images near
public EVCS. The computer-vision technique was used to extract scene features
from panorama images. We conducted a spatiotemporal analysis of public EVCS
usage. The built-environment factors underwent dimensionality reduction and were
assessed for outliers. Descriptive analysis revealed afternoon peak charging times
and variations between chargers. Additional peaks are observed in the weekday late
evening for chargers located near mega-retail stores. Public EVCS in Seoul were
utilized more on weekdays than on weekends. Public EVCS in central business dis-
tricts saw the most significant usage, with potential cases of overuse. An analysis of
the built environment around the chargers showed unique characteristics, with some
forming identifiable clusters. The most used public EVCS had more parking areas
than other POIs, matching visual observations. Computer visioning mainly recog-
nized highways, parking lots, and crosswalks as common features near the chargers.
Outlier test results generally defined fast chargers in the central business district area
as outliers. The results also demonstrated that built-environment measures from POI
data and computer vision can be used in a complementary manner. Our study offers
empirical findings to enhance the understanding of public EV charging usage. We
demonstrated the use of POI data and computer-vision techniques to quantify the
built environment.
Keywords EV· EV charging· Built environment· POI· Computer vision
* Seung Jun Choi
jun.choi@utexas.edu
1 Urban Information Lab, School ofArchitecture, The University ofTexas atAustin, Austin,
TX78712, USA
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
ResearchGate has not been able to resolve any citations for this publication.
Toward an equitable transportation electrification plan: Measuring public electric vehicle charging station access disparities in Austin, Texas
Article
Full-text available
The deployment of public electric vehicle charging stations (EVCS) is a critical component of transportation electrification. Recent studies have highlighted growing concerns about disparities in accessibility to public chargers between different demographic groups. This research expands ongoing equity concerns by contextualizing existing transportation equity discourse and analyzing public charger access disparities in Austin, Texas. Using threshold equity toolkits, we investigated public EVCS access disparity across different races and income groups. We conducted a generalized additive model regression to measure and visualize the effects of possible determinants on public EVCS access. The analysis results revealed that a public EVCS access disparity exists in Austin, with most chargers being installed in areas where the majority of the population is Non-Hispanic White. There was a more equal distribution of public EVCSs across income quartiles when compared with race. However, middle- and high-income groups had better access than lower-income communities in terms of distance to the nearest public EVCSs. Our regression analysis found that regional and socio-demographic factors, such as race and income, have a statistically significant impact on public charger access. The regression analysis also revealed that Austin’s current public EVCS deployment seems to favor communities above the poverty level and with higher numbers of registered electric vehicles. Local policymakers should reflect on the findings of this study to develop an equitable transportation electrification plan. Federal environmental justice plans such as the Justice40 initiative can benefit from incorporating more local contexts to better invest in disadvantaged communities.
View
Uncovering electric vehicle ownership disparities using K-means clustering analysis: A case study of Austin, Texas
Article
Full-text available
  • Jul 2024
Transportation electrification is promoted for its environmental and energy efficiency benefits. However, recent studies examining electric vehicle (EV) adoption have revealed complex patterns influenced by race and income disparities. These studies, primarily based on surveys, often overlook regional ownership variations and built environment measures linked to urban form. Our study addresses this gap by analyzing actual EV registration data with spatial details using hot spot and K-means clustering analyses. The analysis results revealed a pronounced East–West divide in EV adoption. In West Austin, clusters indicate a higher number of EVs, greater energy consumption, and residents who are predominantly White, with higher income and education levels. They mainly live in single-family housing units. Conversely, in East Austin, clusters show a lower number of EVs. They are predominantly home to African-American and Hispanic populations with lower income and education levels, often residing in mobile homes. Land use conditions, such as the availability of green open spaces, play a significant role in this divide. Density, diversity, and design measures of the built environment are lower in East Austin compared to West Austin. We argue that survey-reported preferences for EVs do not always align with actual market behavior. While the 30–45 age group may show a higher willingness to purchase EVs, this interest is not consistently reflected in the actual ownership patterns. Factors like residential choice and the built environment may influence EV adoption rates. A broader set of studies is needed to link urban forms with equity.
View
Advances and pitfalls in measuring transportation equity
Article
Full-text available
  • Jan 2024
  • TRANSPORTATION
Transportation systems play a pivotal role in facilitating access to out-of-home activities, enabling participation in various aspects of social life. But because of budgetary and physical limitations, they cannot provide equal access everywhere; inevitably, some locations will be better served than others. This realization gives rise to two fundamental concerns in transportation equity research and practice: (1) accessibility inequality and (2) accessibility poverty. Accessibility inequalities may rise to the level of injustice when some socioeconomic groups systematically have lower access to opportunities than others. Accessibility poverty occurs when people are unable to meet their daily needs and live a dignified and fulfilling life because of a lack of access to essential services and opportunities. In this paper, we review two of the most widely used approaches for evaluating transport justice concerns related to accessibility inequality and accessibility poverty: Gini coefficients/Lorenz curves and needs-gap/transit desert approaches, respectively. We discuss how their theoretical underpinnings are inconsistent with egalitarian and sufficientarian perspectives in transport justice and show how the underlying assumptions of these methods and their applications found in the transportation equity literature embody many previously unacknowledged limitations that severely limit their utility. We substantiate these concerns by analysing the equity impacts of Covid-19-related service cuts undertaken in Washington, D.C. during 2020. The paper also discusses how alternative methods for measuring transportation equity both better comport with the known impacts of such changes and are consistent with underlying moral concerns.
View
View
Public charging choices of electric vehicle users: A review and conceptual framework
Article
Full-text available
  • Aug 2023
  • TRANSPORT RES D-TR E
Studying electric-vehicle public-charging choices is an important aspect of accelerating electric-vehicle adoption. Understanding which factors would entice existing and potential users to charge their electric vehicles at public locations would provide important evidence for policy-makers, vehicle manufacturers and charging providers. The academic and grey literature has now matured to a point where a critical synthesis of the current state of knowledge is necessary. To fulfil this, this review provides a synthesis and critical discussion of the most up-to-date evidence on public charging choices based on which a conceptual framework to match choices and their determinants is devised. Research gaps and further empirical evidence required in this area, as these emerged from this review, include better understanding of the temporal patterns of public charging, users' preferences for different types of charging locations, payment models (monthly subscription, pay as you go) and payment methods.
View
The Pathway of Urban Planning AI: From Planning Support to Plan-Making
Article
Full-text available
  • Jun 2023
  • J PLAN EDUC RES
Artificial intelligence (AI) is rapidly gaining prominence as a crucial technology to transform and reshape the field of urban planning. However, several unanswered questions persist regarding the potential impacts of AI on urban and regional planning research and practice, as well as the issues involved and the appropriate responses and plans. This paper aims to address these concerns in the AI-enabled planning process and accordingly create a typology of urban planning AI to categorize and outline the progression of AI in urban planning, ranging from AI-assisted and AI-augmented planning to AI-automated and eventually AI-autonomized planning, based on a scoping literature review.
View
Evaluating Air Quality Status in Chicago: Application of Street View Imagery and Urban Climate Sensors
Article
Full-text available
  • Apr 2023
  • ENVIRON MODEL ASSESS
Cities worldwide have initiated the installation of urban climate sensors to monitor air quality in real time and take proactive measures against the growing threat of climate change. This study focuses on the city of Chicago and utilizes Microsoft’s recently launched Project Eclipse sensors to evaluate air quality status. We extracted surrounding land use features near the installed sensors, integrating street view images from Google Street View (GSV) with conventional land use extraction toolkits. Principal component analysis (PCA) was conducted to decompose spatial information and evaluate the unique characteristics of using street view imagery. Additionally, we integrated XGBoost machine learning regression analysis and SHapley Additive exPlanations (SHAP) value calculation to investigate the impact of determinants on air quality. Analysis results indicated that the measured air pollution exposure from the sensors was consistent with the city’s predefined values, except for the Northern intersection of West Belmont Avenue, which reported critical air quality issues. The regression analysis and SHAP calculation revealed significant differences in the impact of land use on air quality between the intersection of West Belmont Avenue and random observations. The city and local government agencies should address the existing built environment and land use conditions in the North to mitigate potential harm.
View
Quantitative land price analysis via computer vision from street view images
Article
Full-text available
  • Apr 2023
  • ENG APPL ARTIF INTEL
Land price is an important economic factor in producing meaningful references for regional planners by assisting them in urban planning, economic decision-making, and land resource allocation. However, related studies in land price analysis were mainly focused on the factors of site area and plot ratio, analysis of the potential impact of streetscape factors and human subjective perception on the land price has been lacking, regardless of the impact on supply and demand relationship. Therefore, this study developed a new approach for estimating and analyzing land prices through deep learning that considered the streetscape and human subjective perception factors. In the estimation part, we developed a fine-grained end-to-end deep learning model, the input is street view images, and the output is land prices. In the analysis part, we extracted the semantic segmentation results and human subjective perception scores from the images and combined them with the results of land price estimation. We then introduced a combination of quantitative analysis using the gradient-weighted class activation mapping and L1-based sparse linear regression to model the relationship between the streetscape and the human subjective perception quantitatively. The gradient-weighted class activation mapping was used to determine which categories of pixels deep learning relied on to output the results of land price estimation quantitatively. Combined with segmentation results, we implemented L1-based sparse linear regression and quantitatively determined the importance of the streetscape factors for land prices. Overall, our deep learning model achieved 77.99 % accuracy by only using street views in estimating the land price, and we illustrated the impacts of the streetscape and perception on the land price by showing that perception scores are more important than streetscape factors such as road and mountain in the streetscape. Comfortable, lived-in feel in perception have the most important impact on land price estimation.
View
View
View