Impacts of ride-hailing services on travel behavior

Hossain Shubho, Md Tanvir

Permalink

https://hdl.handle.net/2142/115517 Copy

Description

Title

Impacts of ride-hailing services on travel behavior

Author(s)

Hossain Shubho, Md Tanvir

Issue Date

2022-03-21

Director of Research (if dissertation) or Advisor (if thesis)

Lee, Bumsoo

Doctoral Committee Chair(s)

Lee, Bumsoo

Committee Member(s)

• Chakraborty, Arnab
• Braun, Lindsay M
• Cidel, Julie L

Department of Study

Urban & Regional Planning

Discipline

Regional Planning

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Ride-hailing
• Travel behavior
• Public Transit
• Car ownership
• Path analysis
• Multinomial Logit Model
• Copula-based Regression
• Joint Decission

Abstract

The rapid development of ride-hailing (RH) services has altered existing modes of transportation throughout the world. The current body of literature is predominately concerned with the questions of whether and to what extent RH services will replace public transport and private automobile usage. Most studies looking into the effect of RH services on public transportation are restricted to an aggregate analysis of metropolitan areas only. The few studies that have looked at the relationship between vehicle ownership and the availability of RH services have treated either RH use or vehicle ownership as a dependent variable and the other as an exogenous variable. These studies ignored the possibility of a relationship between the two occurring simultaneously, potentially leading to biased results. This dissertation addresses these existing research gaps in the study of RH services and their interaction with travel behavior using disaggregated analyses. This dissertation comprises three papers. The first paper investigated the complex relationship between RH service and public transit use among young adults using the 2017 NHTS data. Employing a path analysis, it particularly examined the mediating effect of car ownership and moderating effect of urban form. A descriptive analysis result shows that RH use is negatively associated with both car ownership and vehicle miles traveled (VMT), and the negative correlation is stronger in neighborhoods that are more compact with high density and mixed land use. It seems that RH services enable the carless lifestyle of young adults, supplementing public transit, especially in compact urban neighborhoods that are better served by public transit. Path model results confirm the positive association between RH services and transit ridership mediated through lower car ownership. After controlling for other covariates (socioeconomic variables), the length of time since the introduction of RH services was negatively associated with car ownership in all five quantile samples in terms of a compactness score, and the association was strongest in top quantile. A complete sample model also shows a similar result—the association of RH services with lower car ownership and hence more frequent transit use is stronger in more compactly developed neighborhoods. We also found a similar mediating effect of car ownership and moderating effect of urban form in a path model for vehicle miles traveled (VMT). The findings of this chapter indicate that the availability of RH services has the potential to promote transit ridership and lower VMT through lower car ownership, especially in compact urban neighborhoods. The second paper examined the extent to which RH services are replacing driving and transit trips, using individual trip data from the Chicago metropolitan area in 2019. Unlike existing studies that primarily rely on an aggregate level analysis or survey responses of RH users, this study investigates the interactions of RH services based on actual individual-level mode choice behavior. It conducted a multinomial logit (MNL) analysis to model the determinants of mode choice between driving and transit use, including mode-specific trip time and cost, and traveler’s characteristics. Using the estimated MNL model coefficients for driving and transit trips, we predicted the likely choice between the two modes for the trips that were actually made by RH if RH had not been available. The results show that about 60 percent of RH trips would have been made by transit if RH services had not existed, while 40 percent of RH trips are replacing trips by car. The extent to which RH services are replacing transit versus personal vehicle trips varies across diverse socioeconomic groups and neighborhoods. RH services are replacing more transit trips in high-density neighborhoods, indicating that such services are more likely to compete with public transportation in densely populated areas than in other locations. The third study examined the relationship between RH use and household vehicle ownership in the context of the Chicago metropolitan area. Previous research has produced inconsistent results largely due to data and methodological limitations. This study contributes to the literature by investigating joint decisions of ride-hailing usage and vehicle ownership, using a copula-based regression analysis with individual trip level data. The results show that when a User chooses both vehicle ownership and RH use, these choices have a negative dependency (or discordance) in the Chicago metropolitan area. In other words, the availability of RH services in urban areas such as Chicago may result in a reduction in the number of cars on the road. Results also show that the compactness of a city has a significant influence on the dependence structure between car ownership and RH usage; that is, an increase in RH usage is more likely to result in a reduction of car ownership in densely populated areas. This dissertation contributes to a growing body of empirical evidence on the interaction of ride-hailing, public transportation, and private vehicles in the United States and other countries. In light of the findings from this research, future studies could identify and address gaps in the public transportation network. At the same time, spatial disparities in service provision could give context to some individuals’ preference for RH over public transit in certain parts of cities. Additionally, future research could focus on the distinctions between private and pooled RH services (e.g., UberPool). Such research, which could be conducted using empirical data or matching algorithms, could shed light on the relative positioning of shared rides in relation to private car rides and public transportation. Additional analysis may examine the demand for modes of transportation other than driving and public transit, such as biking or walking, as well as the role of land-use distribution and socioeconomic data, in order to explain the underlying determinants of observed differences in public transportation and ride-hailing use, as well as demand for each.

Graduation Semester

2022-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Md Tanvir Hossain Shubho

Owning Collections