2021

Zheyan Chen (z.chen1@uu.nl), Universiteit Utrecht
Dea van Lierop, Universiteit Utrecht
Dick Ettema, Universiteit Utrecht

The emergence and popularity of dockless bike-sharing systems have attracted extensive attention due to the associated environmental and health benefits. However, little consideration has been given to the potential individual social implications of dockless bike-sharing. Our knowledge about whether dockless bike-sharing systems have the ability to facilitate individuals’ engagement in daily activities is limited. This goal of this study is to gain more insight into how individuals’ personal characteristics and neighborhood environment features influence perceived access to different types of activities by dockless bike-sharing. Using survey data collected from residents in Beijing, we employed four ordinary least squares (OLS) regressions to assess the effect of individual and spatial attributes on the role dockless bike-sharing plays in users’ perceived accessibility to activities overall as well as to three different categories of activities—subsistence, maintenance and leisure. The results indicated that male users reported enjoying more benefits in accessing activities. Dockless bike-sharing users’ perceived benefits in accessing activities largely relied on their social support from their family and friends and their attitudes towards environmental and health concerns of travel. Additionally, users who agreed that dockless bike-sharing has helped them access bus stops and metro stations perceived higher benefits of dockless bike-sharing on activity participation. Our analysis also highlighted that dockless bike-sharing users in Beijing benefited most in their commuting trips, and to a lesser degree, when attending maintenance and leisure activities. The percentage of cycling paths within the home neighborhood tended to be positively associated with individuals’ perceived accessibility to subsistence activities.

Rick Liu (rick.liu@mail.utoronto.ca), University of Toronto
Matthew Palm, Worcester State University
Steven Farber, University of Toronto, Scarborough
Amer Shalaby, University of Toronto

Two common options for commuters stranded during subway disruptions are waiting for transit agency provided bus-bridging service or switching to ride-hailing (Uber/Lyft). Many agencies around the world use active buses to provide bus-bridging services during subway disruptions, leading to reduced service for bus rides. These bus-bridging policies and the use of ride-hailing raise questions regarding the equitable distribution of capacity during subway disruptions. After analyzing 78 subway disruptions in Toronto, we found that more vehicles used for bus-bridging services came from routes serving disadvantaged populations. In most cases, routes serving disadvantaged populations were affected worse than routes not serving disadvantaged populations. Ride-hailing pickups in subway disruptions at disadvantaged areas in Toronto were less concentrated than other areas of the city, and produced fewer hotspots of ride-hailing activity, which could either indicate a service disparity or a demand disparity resulting from the different demographics able or willing to use ride-hailing.

Jane Gould (info@janegould.com), WAVA

Ridesource services like Uber and Lyft could be a vital transportation option for older people. During the Covid crisis, many seniors have been reluctant to use public transportation. Others, who reside in low-density suburbs, shun taxis and seek an alternative when they are no longer able to drive. Despite the appeal of ridesource, there are very few seniors who have used it. This study explores the reasons, and centers on two factors: first, lack of knowledge and experience with smartphones, and second, fear or anxiety about this new mode of travel. From a pre-class survey we identify three related constructs: age challenges, technology, and the centrality of cars in elder’s daily lives. We demonstrate that travel training can play an important role and assist with the digital divide. In this study, we provide survey results from 115 seniors that signed up for a ridesource travel training class. We explore their expectations and beliefs prior to taking their first trip. We also report on survey data from 26 seniors who took the class, and then received travel vouchers for rides on Lyft over a three-month period. Almost all the participants found that they could manipulate the app and said they felt comfortable using it. They also said that they would recommend ridesource to peers. We discuss ridesourcing as the first step on a ladder that can accelerate digital learning and interactions for older people, and make transportation more accessible.   Keywords: Ridesource, Digital Equity, Travel Training, Technology, Uber, Lyft, Ridehailing, Accessibility, Smartphones

Xinyi Wu (wuxx1088@umn.edu), University of Minnesota, Twin Cities
Xinyu (Jason) Cao, University of Minnesota
Frank Douma, University of Minnesota

As an emerging technological innovation, vehicle automation will have profound impacts on various aspects of society.  Despite many initial discussions on related topics, few of the studies have emphasized how this new technology will affect equity. To better understand how autonomous vehicles (AV) might affect equity, this study explores the potential influence of AV on eight groups of the transportation-disadvantaged population. Based on existing travel behaviors of the identified population, we conclude that AVs tend to bring more benefits to some people, but might have mixed effects on others. Based on our findings, we further provide implications for future policymaking.

David Duran (david.duran@tum.de), Technical University of Munich
Francisco Camara Pereira, Technical University of Denmark
Gebhard Wulfhorst, Technical University of Munich

Bike-sharing systems (BSS) are known to provide health, environmental, social, and financial benefits. These benefits, however, have not been usually equally distributed among the population with an over-representation of high income and highly educated male, young, and white people.  This unequal distribution of benefits is perpetuated when the main goal of BSS is efficiency, as station locations and service areas are prioritized in areas where this segment of the population lives and commutes. Therefore, we developed a heuristic and data-mining-based method (DARE) to weight efficiency and equity in the planning process of BSS. The primary inputs are the predicted potential demand (efficiency) and a spatial deprivation index (equity).  Potential demand is predicted using structural equation models(SEM) to help understand the relationship between predictors as well as test and validate the theoretical assumptions. DARE was applied to a hybrid BSS in Munich, Germany to evaluate the distribution of stations following spatial efficiency and equity criteria. The design was in line with recommendations from guidelines and findings from previous research when spatial efficiency instead of equity was considered in the application. On the other hand, the neediest population was prioritized when spatial equity was considered but central areas were excluded. System designers have now available a decision-making method where fairness is part of the input on the planning process for BSS. Moreover, through this method, the public can know who was prioritized during the planning of a system: potential demand, the population most in need, or both of them.

Xinyi Wang (xinyi.wang@gatech.edu), Georgia Institute of Technology (Georgia Tech)
Xiang Yan, University of Florida
Xilei Zhao, University of Florida
Zhuoxuan Cao, University of Florida

Concepts of Mobility-on-Demand (MOD) and Mobility as a Service (MaaS), which feature the integration of various shared-use mobility options, have gained widespread popularity in recent years. While these concepts promise great benefits to travelers, their heavy reliance on technology raises equity concerns as socially disadvantaged population groups can be left out in an era of on-demand mobility. This paper investigates the potential uptake of MOD transit services (integrated fixed-route and on-demand services) among travelers living in low-resourced communities. Specially, we analyze people's latent shared mobility preference towards three shared-use mobility services, including ride-hailing services, fixed-route transit, and MOD transit. We apply latent class cluster analysis to 825 survey respondents sampled from low-resourced neighborhoods in Detroit and Ypsilanti, Michigan. We identified three latent segments:  shared-mode enthusiast ,  shared-mode opponent , and  fixed-route transit loyalist . People from the  shared-mode enthusiast  segment often use ride-hailing services and live in poor-transit-access areas, and they are likely to be the early adopters of MOD transit services. The  shared-mode opponent  segment contains a large proportion of vehicle owners who lack interests in shared mobility options. The  fixed-route transit loyalist  segment includes a large proportion of low-income individuals who face technological barriers to use the MOD transit. We also find that being males, having a college degree, owning vehicles, having mobile data plans, and living in poor-transit-access areas are associated with a higher level of preferences for MOD transit services. We conclude with policy recommendations for the development of more accessible and equitable MOD transit services.

Xilei Zhao (xilei.zhao@essie.ufl.edu), University of Florida
Xinyi Wang, Georgia Institute of Technology (Georgia Tech)
Xiang Yan, University of Florida
Zhuoxuan Cao, University of Florida

The future of public transit service is often envisioned as Mobility-on-Demand (MOD), i.e., a system that integrates fixed routes and shared on-demand shuttles. The MOD transit system has the potential to provide better transit service to all individuals, especially the disadvantaged populations. However, little research has focused on understanding traveler preferences for MOD transit and the preference heterogeneity across population groups. This study addresses this gap by proposing a latent segmentation based decision tree (LSDT) method. This method first uses a latent class cluster analysis (LCCA) that extract traveler profiles who have similar usage patterns for shared modes. Then, decision trees (DT) are adopted to reveal the associations between various factors with preferences for MOD transit across different clusters. We collected stated-preference data among two low-resource communities, i.e., Detroit and Ypsilanti, Michigan. The LCCA model divides the entire sample into three clusters, i.e., shared-mode users, shared-mode non-users, and transit-only users. We find that job accessibility by transit is the most important variable for all the cluster-specific DT models. For transit-only users, gender and car ownership are the second-important variables, but neither of them appears in the DT for the other two clusters. In particular, female transit-only users have lower preference for MOD transit, possibly due to safety concerns. The LSDT method can generate richer insights than a single DT fitted to the overall sample. The insights gained from this approach can generate valuable insights and inform better-targeted strategies to plan for MOD transit services.

Shuocheng Guo, University of Alabama
Eleftheria Kontou (kontou@illinois.edu), University of Illinois, Urbana-Champaign

The path towards light-duty vehicles electrification promises benefits like lower operating costs for drivers and reduced environmental externalities for all. Policies such as electric vehicle rebates assist with alleviating the high capital costs of alternative fuel cars compared to conventional ones. Such incentives though have not been uniformly allocated to consumers of different income levels. We uncover distributional effects of plug-in electric vehicle rebates in California. We leverage economic attributes representative of populations of Californian census tracts as well as the number and amount of rebates distributed to plug-in electric vehicle buyers through the Clean Vehicle Rebate Project from 2010 to 2018. Horizontal and vertical equity measures are evaluated using Gini and Suits coefficients. Local measurement of spatial autocorrelation characterizes spatial patterns of rebates allocation across the state of California. We evaluate the rebate allocation geographical and distributional fairness across income groups and disadvantaged communities (DACs). We find that rebates have been predominantly allocated to higher-income PEV buyers. However, the share of rebates distributed to low-income groups and DACs increased after an income cap policy was put into effect. Spatial analysis shows high spatial clustering effects and rebates concentration in major metropolitan regions such as Los Angeles, San Francisco, and San Diego. We reveal neighborhood effects showing that communities with lower median income that are disadvantaged receive a greater share of rebates when these are close to spatial clusters characterized as high income and high rebates amount receivers.

Tina Tang, University of Virginia
Mohamad Alipour, University of California, Los Angeles
Amanda Poncy, City of Charlottesville
Devin Harris, University of Virginia

While the potential benefit e-scooters offer to communities to address short distance transportation needs have been shown, evidence that those needs are being met in practice is mixed. This study evaluates whether e-scooters are available to residents that are more likely to need alternative transportataion. This paper presents a geospatial analysis approach investigating e-scooter use conditioned on e-scooter availability to expand upon the micromobility literature.  The demand distribution for a short-distance transportation alternative was visualized in the city of Charlottesville, Virginia based on Census statistics describing transportation use, highlighting areas where micromobility need is the most salient. Real-time e-scooter GPS data was harvested from an open data feed over a four-month period from March 15, 2020 to July 15, 2020 and subsequently processed into measures of e-scooter availability and utilization at U.S. Census block group level resolution. This e-scooter data was then fused with demographic and built environment data and a multiple regression analysis approach was used to model average e-scooter availability and average fleet utilization to investigate which short-distance transportation need factors, economic activity factors, and built environment factors drive the respective response variables. Findings suggest that e-scooter fleet distribution is highly influenced by centers of economic activity while the e-scooter usage is influenced by indicators of residential micromobility needs. Further, this study suggests that e-scooter utilization could serve as a metric for reevaluating e-scooter placement to optimize residents’ needs which, if effective, may also lead to increased e-scooter benefits to a city’s residents.

Samuel Speroni (ssperoni@ucla.edu), University of California, Los Angeles

The Every Student Succeeds Act (2015) gave foster youth additional legal protections in school, including the right to transportation and the right to remain at their school despite any moves, similar to protections already in place for students experiencing homelessness and students with disabilities.  California’s compliance with this mandate was relatively more difficult than other states’, as only eight percent of students in California travel by school bus, compared with 35 percent nationally.  Thus, California schools could not simply tap into these existing services to provide transportation for foster youth. Ridehailing offers a solution to this gap.  HopSkipDrive, a ridehailing company designed to transport children, engages in contracts with school districts and counties to provide school transportation for these vulnerable student populations.  In 2018–2019, HopSkipDrive provided 32,796 trips to school in Los Angeles County, with massive time savings over the logical alterative: transit.  HopSkipDrive offers time savings of nearly 70 percent compared with the same trips simulated on transit. HopSkipDrive’s trips average 28 minutes in duration, yet on transit only 30 percent would have taken less than 45 minutes.  This is despite 90 percent of all origins and destinations being located within a half-mile of a transit stop.  This service has important social equity implications beyond just time savings offered to vulnerable student populations, as HopSkipDrive contract trips tend to originate in neighborhoods with high percentages of low-income households and people of color.

Jeffrey Jobe, University of Texas, San Antonio
Greg Griffin, University of Texas, San Antonio

As the COVID-19 pandemic continues across the world, many elements of normal life are changing. The newly developing isolated lifestyle many are now adopting is reshaping aspects of transportation. Bike sharing is a shared-use micro-mobility system that offers an alternate form of socially distant transportation to citizens for both commuting and leisure. This paper seeks to address concerns associated with bike sharing during the COVID-19 pandemic. Survey results in San Antonio, Texas, suggests that most, but not all, users continue bike share use during the pandemic, and most will increase once restrictions are lifted. Additional evidence also indicated that cycling has increased in many cities throughout the United States. With the possibility of cycling increasing, it is imperative to implement proper precautions regarding the health and safety of bike share users. Further, over half of survey respondents were unaware of precautionary policies, signaling a need to improve communication of practices broader in the community. Extreme sampling of responses by income level shows that low-income bike share users emphasize practical issues such as the time limits for trips before accruing additional charges. In contrast, high-income users highlighted desires for expanding the system into new neighborhoods. Findings of this investigation, results of the San Antonio Bike Share Survey, and research conducted on COVID-19’s impact on transportation provided six recommendations for bike share operators. Bike share operators should continue services while implementing appropriate guidelines for its users and ensure compliance by informing users of guidelines.

Anaka Maher, University of California, Berkeley
Carol Atkinson-Palombo, University of Connecticut
Norman Garrick, University of Connecticut

Ridesourcing services such as Uber and Lyft have seen a large growth in New York City’s outer boroughs. As an on-demand service, ridesourcing has potential to fill gaps in areas where service by public transit is poor and residents also have limited access to personal automobiles. New York City requires ridesourcing services to report trip origin and destination information to its Taxi and Limousine Commission which then makes these data publicly available. Analysis of trip data shows a shift in usage patterns between 2015 and 2018 and distinct time-of-day and day-of-the-week patterns for different types of neighborhoods in the study area in 2018. Outer borough neighborhoods show heavier usage patterns during the morning and afternoon commuting times than their Manhattan neighbors. GIS-based analysis also shows a strong “distance decay” effect with the highest percentage of trips localized within the same taxi zone and tapering off sharply in farther zones. The results point to ridesourcing increasingly being used for commuting in low-income minority neighborhoods. If this essential travel is being relegated to ridesourcing services because of a lack of other viable options, this becomes a transportation equity issue as private ridesourcing companies are largely unregulated and currently functioning with unsustainable pricing structures. With the use of these services still evolving, changes in service could disproportionately negatively affect these already underserved communities.