2021

Long Cheng (long.cheng@ugent.be), Ghent University
Kunbo Shi, Universiteit Gent
Jonas De Vos, University College London
Mengqiu Cao, University of Westminster
Xuewu Chen, Southeast University
Frank Witlox, Universiteit Gent

Understanding the relationship between the built environment and walking among older adults could offer important insights into land use and transport policies which seek to promote active ageing. However, most previous studies have explored global relationships, i.e. the effects are averaged or assumed to be constant over the region of interest. In this study, we focus on the local spatial variations in the relationship between the built environment and the daily time spent walking by older adults. We apply a geographically weighted regression (GWR) model, using data collected from 702 older adults in Nanjing, China. Our results show that spatial heterogeneity exists for built environment effects within the entire study area. It has an impact on all the relationships, with nuances in the significance level, parameter magnitude or sign reversals, depending on the location. Therefore, policy interventions would only be effective in certain areas for certain built environment attributes. By exploring the local contexts of relationships, we further suggest that the spatial heterogeneity stems from contextual effects, i.e. the specificities of places with a discriminative composition of individual and/or environmental characteristics. Our findings can help to enrich the understanding of associations between land use and travel behaviour, as well as offer local planning guidance for creating age-friendly neighbourhoods.

Yudai Kamada, Chuo University
Masayoshi Tanishita, Chuo University

Fuel tax revenue has been decreasing due to technological improvements of vehicles. Road user charge (RUC) including vehicle miles traveled (VMT) fee has been discussed for bearing the increasing road maintenance cost and reducing negative externalities. In this paper, we estimated a gasoline price elasticity on car mileage in Japan using more than 30,000 self-reported panel samples. We applied for the latent class regression analysis considering the reliability of each sample.  As a result, about 30% of households with a long mileage who lives in areas where public transportation is inconvenient, had a significant gasoline price elasticity (-0.2). In Japan, the estimated gasoline price elasticity was –0.09.

Juan Nicolás Gonzalez (juannicolas.gonzalez@upm.es), Universidad Politécnica de Madrid
Jose Perez-Doval, Universidad Politecnica de Madrid
Juan Gomez, Universidad Politécnica de Madrid
Jose Manuel Vassallo, Universidad Politecnica de Madrid

City characteristics influence the use of private vehicles, and the decision to purchase a vehicle, leading to an increase of transport externalities. Policymakers are devoted to reducing externalities through transport policies. This research explores to what extent car ownership is influenced by the implementation of transport policies aimed at improving urban sustainability in the city of Madrid (Spain). By the estimation of a Multilevel Ordered Logit Model, the behavior of the number of household cars were analyzed according to household socio-demographic characteristics, city built-environment variables, transport network attributes, and policy-related variables. The results indicate that City-built environment variables explained car ownership trends in households. Also, the current policies implemented in the city make the use of private cars less attractive, thus impacting negatively on residents’ choices to purchase or own a car. The analysis also assesses the relationship between transport policies and levels of income.

Mairin McKnight-Slottee, University of Washington
Chang-Hee Christine Bae, University of Washington
Edward McCormack, University of Washington

The central theme of U.S. transportation planning policies has shifted in recent decades to include a focus on accessibility, sustainability, and the efficiency gained by strategic coordination with land use planning. To reduce single-occupancy vehicle trips and promote transit and non-motorized transportation, cities often implement Transportation Demand Management (TDM) programs intervening with new development during the municipal permit and design review process. In Seattle, one program involves a commitment from developers to implement select strategies from six TDM element categories: physical improvements, bicycle/walking programs, employer-based incentives, transit, car/vanpooling, and parking management. Under a joint Director’s Rule (DR) from the Department of Transportation and the Department of Construction and Inspections, the Transportation Management Program (TMP) targets new developments and requires some TDM elements, recommends others, and leaves to negotiation the rest. The result is an individualized TMP contract that is site-specific, reflecting both City policy and developer needs. This paper presents a qualitative analysis of the guiding DRs and TMP contracts implemented in Seattle’s Downtown and adjacent South Lake Union area since 1988. A content analysis of TMP documents reveals that the average number of required elements ultimately implemented in a contract falls short of expectations set by DRs. Despite differences in development patterns and transportation resources, the results reveal limited consistency or patterned response in implementation of TDM elements by category between cases. However, the opportunities exist to structure the guidelines in DRs to better leverage developer self-interest in transportation management to align more closely with City policy.

Sicheng Wang (sicheng.wang@rutgers.edu), Rutgers University, New Brunswick
Robert Noland, Rutgers University

Transportation network companies (TNCs) offer a ride-splitting option for ridesourcing trips, allowing users to share the vehicle with others at a lower fare. While encouraging shared rides has environmental benefits, little is known about the price effect on users’ decision to choose the ride-splitting service. Using TNC trip data from Chicago, we investigate the temporal and spatial distribution of authorized ride-splitting trips in 2019. We found choosing ride-splitting can save from $1.57 to $2.13 per mile on average over the year. The willingness to share TNC trips was related to the price difference between shared and non-shared trips. The willingness to share also differed across neighborhoods with different demographics, socioeconomic status, and built environment characteristics. We perform logistic regression and random forest models to determine the marginal price effect on the decision to share. The results indicate the probability of authorizing a ride-splitting trip was highly elastic to the price per mile – the most important predictor in the random forest prediction, which had better accuracy than the logistic model. Policy implications for increasing shared trips are discussed based on the findings.

Huijun Tan (huijutan@pdx.edu), Portland State University
Nathan McNeil, Portland State University
John MacArthur, Portland State University
Kelly Rodgers, Portland State University

This study looked at initial results from the Transportation Wallet for Residents of Affordable Housing (TWRAH) pilot program launched by the City of Portland’s Bureau of Transportation (PBOT). The program provided a set of transportation incentives for low-income participants including a $308 pre-paid US Bank visa card which could be applied to public transit or other transportation services, a streetcar pass, and free bikeshare membership. In September 2019 the program kicked-off at 7 affordable housing locations. We conducted a survey with the program’s participants (243 total responses) to understand how they used the Transportation Wallet and how the program helped them use different modes to get around. The main findings include: 1) The financial support of this program encouraged participants to use new mobility (including Uber/Lyft, BIKETOWN, and e-scooter) that they had never used before; 2) TWRAH program helped participants to use various transportation services, and get to more places otherwise they could not; 3) Transportation Fairs promoted both mode sign-up and mode usage, particularly for new mobility and the TriMet reduced fare program. The survey results also reveal some possibilities to improve the program. Transportation agencies can consider apply more simple or understandable methods/strategies to educate participants how to use new mobility and coordinate (or inform) different service providers better to optimize seamless services for participants. The paper provides some insights into the implementation and effectiveness of a transportation program with financial incentives on how it helps low-income populations.

Kexin Geng (14241238@bjtu.edu.cn), Beijing Jiaotong University
Devi Brands, Vrije Universiteit, Amsterdam
Erik Verhoef, Vrije Universiteit, Amsterdam
Yacan Wang, Beijing Jiaotong University

Tradable permits have received growing attention in the transportation literature as a policy alternative to road pricing. However, empirical evidence of participants' responses to the tradable permits scheme is still limited. This study contributed to provide the first real-life evidence of tradable permits to manage rush-hour travel behavior. By conducting a two-week tradable permit experiment among 91 first-year students in Beijing, this research collected the revealed preference data of participants' time choices and transactions in a web-based permit market. The results show that the tradable permit scheme reduces about 20% of peak trips. Nested logit models revealed that the tradable permits scheme had a significant influence on time choices, and participants with different characteristics showed heterogeneous responses. Behavioral biases also occurred when using the tradable permits scheme. An inequitable valuation for permits and their equivalent market price, and reference dependence using initial permit allocation as the reference point, have been found among our samples. This study empirically informs researchers and policymakers on how the tradable permit scheme performs in reality and provides ground for further experimentation and studies.

Lin Xiao (18811508829@163.com), Tongji University
Jiyan Wu, Tongji University
Jian Sun, Tongji University
Ye Tian, Tongji University

Incentive-based traffic demand management (IBTDM) strategies utilize rewards to redistribute travel demand across space and time. Such congestion alleviation solutions are usually managed by small private companies with constraint budgets. Aside from spending money on incentives, running promotional campaigns to achieve the gains in market share is essential for maintaining the financial health of IBTDM programs. Therefore, the budget allocation between the two counterparts - incentive and marketing expenditure - needs to be wisely determined. Based on the bottleneck model, this paper proposes an optimal budget allocation scheme considering the impact of a budget constraint and market penetration. It was found that the constraint budget should be prioritized to attract those with lower marketing costs in general. In situations with an insufficient budget and when marketing costs were lower for attracting lower-income individuals, IBTDM decision-makers should focus on those lower-income individuals at first. This mitigates inequity issues to some extent.

Weijie Yu, Southeast University
Wei Wang, Southeast University
Xuedong Hua, Southeast University
Xueyan Wei, Southeast University

Accurate analysis of traffic demand in urban is the key to active traffic control and road guidance. Distribution of traffic demand shows its variability and complexity under comprehensive land-use scenarios. Researchers on this field widely study the relationship between traffic demand and land-use scenarios, while the intensity of land-use is ignored when determining land-use scenarios and the distributions of traffic demand in each land-use scenario are not studied specifically. To fill this gap, this paper aims to explore traffic demand distribution of comprehensive land-use scenarios. Traffic analysis zones are generated by clustering origin-destination points of floating taxis. Using POI data, comprehensive land-use scenarios are determined by combining land-use forms and land-use intensity. Then, K-shape algorithm is adopted to extract the typical distributions of traffic demand in each land-use scenario. Finally, total traffic demand (TTD) and traffic demand difference (TDD) are computed and their spatiotemporal characteristics are further analyzed. The results indicate traffic demand distributions are still differentiated even under the same land-use scenario. Three Land-use scenarios with average hourly traffic demand reaching about 300 veh per kilometers have the largest TTD and uneven TDD. This study is potential to provide accurate guidance for relieving traffic congestion caused by disordered urbanization.

Scott Middleton, EBP US
Kyle Schroeckenthaler, EBP
Deepak Gopalakrishna, ICF International Inc
Allen Greenberg, Federal Highway Administration (FHWA)

Transportation network companies (TNCs) offer two types of service: private party ridehailing and shared ridehailing. Policymakers have an interest in encouraging shared over private ridehailing in order to promote more efficient use of the transportation network. While transportation researchers have analyzed ridehailing behavior before, there is limited literature describing the effect of price and time on a rider’s choice between private party and shared ridehailing. This paper fills this gap by analyzing revealed preferences for private party and shared ridehailing trips in 15 American cities coupled with a survey of 4,365 users of a large TNC that includes stated preference questions focused on various alternative options for their most recent trip choice. This study finds that an increase in the relative price difference of $1 per mile increases an individual’s probability of sharing by over 8 percent, while a decrease in the relative travel time difference of 1 minute per mile increases the probability of sharing by over 33 percent. The survey results also show that that a sizeable portion of private party TNC trips (approximately 35 percent) will be difficult or even impossible to convert to shared rides through a price-based incentive. Market segmentation analysis reveals user and trip types where price- and time-based incentives have a relatively greater effect on the choice between private and shared rides. Finally, heterogeneity in user time versus money trade-offs suggests new product possibilities that would increase TNC sharing.

Mashrur Rahman, University of Texas, Austin
Krishna Murthy Gurumurthy, University of Texas, Austin
Kara M. Kockelman (kkockelm@mail.utexas.edu), University of Texas, Austin

Increasing number of corporations and workplaces have begun to provide flexible working hours, or flextime, for employees, which is expected to reduce congestions by redistributing the temporal pattern of commuters’ departure time.  This study examines the impacts of flextime on departure time choice using a Bayesian continuous-time hazard duration model. The model accommodates the time-varying effect of covariates and unobserved heterogeneity. Results from the Austin Household Travel Survey collected between 2017 and 2018 show that workers who have a flextime option choose to leave later, with a predominant effect deterring AM peak departures. Other trip and individual-specific variables such as travelers’ job type, trip duration, number of trips during the travel day and household income were found to have significant impacts on departure time choice. The results also show that flextime is more effective shifting the departure time for retail and service sector employees, those who travel longer and perform more daily activities.   The findings of this study reconfirm the theoretical underpinnings that implementing such policies may ease congestion by staggering the travel demand from peak to off-peak hours.

Darcy Reynard, University of Alberta
Manish Shirgaokar (manish.shirgaokar@ucdenver.edu), University of Colorado, Denver
Damian Collins, University of Alberta

Over the last two decades, we have witnessed the arrival of new forms of travel (e.g., dockless bicycles, e-scooters), social movements (e.g., tactical urbanism, public protest), and most recently the COVID-19 pandemic, which have challenged cities to re-imagine the public right-of-way. These shifts have changed the use and regulation of street space. In this paper, we investigated how North Americans are viewing the alternative uses and forms of street management, specifically during the COVID-19 pandemic. We relied on a corpus of roughly 197,000 geolocated tweets across the U.S. and Canada. Using word vectors, we assigned a numerical value to each tweet, based on its content around three concepts, namely, curbside, mobility, and public space. We depended on a spatial analysis of four metropolises—New York, NY, San Francisco, CA, Toronto, ON, and Vancouver, BC—to study where discussions about curbside were most concentrated geographically from March 18, 2020 to July 1, 2020. We also relied on a deeper qualitative assessment to characterize the conceptual areas the tweets cover. Our analysis suggests that tweeted views about the public right-of-way are concentrated in urban cores. Locations with greater demands on curb space are likely to have the most tweets, positive or negative, about street re-design. The textual content of the tweets indicates that the sidewalk and curb may need to undergo a reassessment given possible emergent needs for extending businesses into the street, adding more capacity for active travel modes, and generating more physical separation between users.

Evan Iacobucci (evan.iacobucci@rutgers.edu), Rutgers University

Motivated by concepts employed in travel habits and mobility biographies research, this project explores the relationship between key life events and shifts in routine transportation behavior. It uses data scraped from Reddit, a popular social media and content-sharing website, to observe real-world conversations about personal travel histories. Specifically, it focuses on understanding the relationship between changes in travel patterns and key events that correspond with these changes. Directed content analysis techniques are used to analyze 437 comments from three US cities: Atlanta, Boston, and Washington, DC.  The results suggest two distinct pathways through which routine travel behaviors change: 1) through influence of a key event, and 2) through reevaluation of available options. In the first pathway, while key events hasten change by making people consider their options, they play a causal role by either altering the transportation choices available to a person or altering their transportation needs. In the second, people reconsider their current patterns and opt to make a change, but these changes appear to happen unprompted by a key event. These results inform a goal of sustainable transport policy in two ways. First, the best way to leverage key events into less car-dependent behavior is to ensure that viable alternatives are present when these events happen. Second, people are capable of noticing and reacting to incremental changes in the quality of available options and will likely respond to them.

Doaa Al-Alawneh, Jordan University of Science and Technology
Anne Gharaibeh, Jordan University of Science and Technology
Jaser Mahasneh, Jordan University of Science and Technology
Ahmad Alomari, Yarmouk University

The primary objective of this research is to provide a comprehensive assessment concerning factors that influence driving alone decisions and parking behaviors among students in rural universities. This research performed quantitative analysis and applied regression models to understand student parking demands and requirements in rural campuses throughout reviewing the most crucial factors that affect driving alone decisions and parking behaviors. A sample of 1252 students at Jordan University of Science and Technology (JUST), Irbid, Jordan, collected through a web-based survey, was used. Results indicated that driving alone decisions for students in rural campuses were determined by a group of significant socioeconomic and psychological variables. Socioeconomic and demographic variables include gender, marital status, age, residential location, travel cost, and number of private vehicles per household. Correspondingly, the ability to find a parking space was heavily influenced by psychological variables and perceptions, including the number of parking days per week, arrival time to students’ parking, time spent searching for a parking space, being late to class while looking for a parking space, and trip chaining. Evaluating these factors is necessary to decrease the dependency of JUST students on private vehicles and to explore the tools that might be used to raise their awareness about using other mode choices. Rural campuses feature large and open spaces away from urban activities, congestion of population, air pollution, and noisy traffic. Comprehensive parking management strategies on rural campuses are substantial since alternative modes of travel are limited due to isolation from the city and settlement areas.

Jason Soria (jason.soria@u.northwestern.edu), Northwestern University
Amanda Stathopoulos, Northwestern University

We expand the literature on the demand for new mobility platforms by exploring spatial dependence in ridesourcing use. This paper employs a Social Disadvantage Index, transit access analysis, and a Spatial Durbin Model to investigate the influence of both local and spatial spillover effects on the demand for shared and solo ridesourcing. We analyse 127 million ridesourcing rides in Chicago from the end of 2018 through 2019. The results show that the effects of built environment and social conditions are similar for shared vs. solo rides, though with a significant distinction. Specifically, we find positive direct and spillover effects in which the concentration of poor social conditions is correlated with higher usage of pooled rides whereas it is correlated with lower usage for solo rides. Additionally, we find that transit access has a substantial, non-intuitive effect in both models; as transit access gets worse (i.e. walking access time to a rail station gets longer) the demand for both solo and shared rides decrease. With these results, we examine the implications of the models on how users view the two modes, how access to shared rides can impact community outcomes, and how transit interacts with ridesourcing.

Jerome Laviolette (jerome.laviolette@gmail.com), Ecole Polytechnique de Montreal
Catherine Morency, Ecole Polytechnique de Montreal
E. Owen Waygood, Ecole Polytechnique de Montreal
Konstadinos Goulias, University of California, Santa Barbara

Car ownership is linked to higher car use, which leads to important environmental, social and health consequences. While car ownership keeps increasing in most countries, it remains relevant to examine what factors and policies can help contain this growth. Most recent analyses of car ownership were conducted using disaggregated models. This paper uses advance spatial econometric modelling framework, to investigate spatial dependence in household car ownership measured at fine geographical scale using administrative data. Such fine resolution allowed for the use of more explanatory variables than previous aggregate models of car ownership. Formal tests confirm the choice of the Spatial Durbin Error Model as the best modeling option. Results indicate that sociodemographic variables explain much of the observed spatial dependence, but that built environment characteristics, including transit level of service and local commercial accessibility also play a role. Yet, the set of included explanatory variables could not explain all the remaining autocorrelation in the linear model residuals, indicating that some important variables are omitted. The SDEM specification allows the interpretation of the model despite these omissions.

Khatun E Zannat, University of Leeds
Charisma Choudhury (cfc@alum.mit.edu), University of Leeds
Stephane Hess, University of Leeds

Dhaka, one of the fastest growing megacities in the world, faces severe traffic congestion leading to a loss of 3.2 million business hours/day (1). While peak-spreading policies hold the promise to reduce the traffic congestion levels, the absence of comprehensive data sources makes it extremely challenging to develop econometric models of departure time choices for Dhaka. This motivates this paper where we develop advanced discrete choice models of departure time choice of car commuters using secondary data sources and quantify how the level-of-service attributes (e.g. travel time), socio-demographic characteristics (e.g. type of job, income, etc.) and situational constraints (e.g. schedule delay) affect their choices.  The trip diary data of commuters making home-to-work and work-to-home trips by personal car/ride-hailing services (572 and 549 respectively) have been used in this regard. Given the discrepancy between the stated travel times and those extracted using the Google direction API, a sub-model is developed first to derive more reliable estimates of travel time throughout the day. A mixed multinomial logit model is then developed to account for the schedule delay from the preferred departure times (PDT, unobserved). Estimation results indicate that the choices are significantly affected by the travel times, activity duration and schedule delay and there is substantial heterogeneity depending on the type of job. The influence of the type of job on PDT was estimated using two different distributions of PDT for the office employees and self-employed people (truncated normal and Johnson’s SB distribution respectively). In addition to being practically useful for devising peak-spreading policies in Dhaka, the proposed framework can be useful in other developing countries with similar data issues.