2020

María Alonso-González, Delft University of Technology
Oded Cats, Delft University of Technology
Niels van Oort, Technische Universiteit Delft
Sascha Hoogendoorn-Lanser, Delft University of Technology
Serge Hoogendoorn, Technische Universiteit Delft

Simulation studies suggest that pooled on-demand services (also referred to as ridesplitting or shared ride-hailing/ridesourcing services) have the potential to bring large benefits to urban areas with little time loss for their users. However, the large majority of on-demand requests in existing services are for individual rides. In this study, we investigate to what extent fare discounts, time losses, and the (un)willingness to share the ride with (different numbers of) other passengers play a role in the decision of individuals to share rides. To this end, we designed a stated preference study, and simulated different scenarios based on an estimated discrete choice Mixed Logit model. We found that the share of individuals who are willing to share their rides depends primarily on the time-cost trade-off they encounter, rather than on the disutility stemming from pooling rides per se. We also found that the disutility of having one or two extra passengers is constant regardless of the trip length, whereas it further increases with trip distance in the event that one shares the ride with four additional passengers. The depicted scenarios show the impacts that larger/smaller time-cost trade-offs and different number of individuals in a ride would have on the shares of individual versus pooled on-demand services. These insights are relevant to on-demand service providers who wish to offer pooled services and to transportation authorities who may be interested in internalising the externalities stemming from the increase in vehicle miles travelled related with individual rides.

Tatsuya Fukushige, University of California, Davis
Susan Handy, University of California, Davis

A modal shift from car to bicycling could produce broad benefits for transportation, energy consumption, environmental quality, public health, and quality of life. While major cities in the US have promoted bicycling through a combination of investments in infrastructure and soft programs, bicycling remains at 1.1% of trips in the US compared to more than 10% in some European countries. This research focuses on individual bicycling propensity and explores differences in characteristics between regular bicyclists, people who do not regularly bicycle but have positive attitudes toward biking, and those who do not have positive attitudes towards biking. The characteristics examined in this paper include socio-demographics, transportation resources, attitudinal factors, and perceptions of transportation modes. The California Millennials Dataset 2015, which collected data on a wide-array of travel behaviors and attitudes for California residents from the Millennial generation and Generation X, is used. The analysis using ordered logit models shows significant associations with student status and gender from among the socio-demographic factors. Current residential location, neighborhood type in childhood, and bike share experience from among transportation resources have significant effects on bicycling propensity. Health and environmental concern lead to higher bicycling propensity. Those who perceive that active transportation is an environmental-friendly mode and enables them to get where they need to go are likely to be in a higher category of bicycling propensity. The results will help policy makers in developing strategies for promoting a modal shift and may help the private sector in developing the next generation of bike-sharing systems.

Matthias Sweet, Ryerson University

Understanding how on-demand ride-hailing and vehicle automation may affect travel behavior is important in order to craft better policy.  This study focuses on the Greater Toronto and Hamilton Area in Ontario, Canada and uses a 2018 survey of 2,923 respondents to explore user interest in these mobility services.  Stated preference choice experiments are administered and mixed logit models are estimated to explore consumer interest in adopting driverless on-demand vehicles, shared on-demand vehicles, transit plus on-demand vehicles, and driverless public transit shuttles.  Monetized generalized costs of each of these choice attributes are estimated for four user groups: commuters and non-commuters each from vehicle-holding and zero-vehicle households.  Results indicate significant generalized costs from combining transit with on-demand ride hailing (a frequently-contemplated “last-mile” solution), while users are relatively ambivalent about shared on-demand ride-hailing.  Results indicate that consumers view driverless public transit shuttles and driverless cars as conferring higher generalized costs than vehicles with drivers.  Overall, results suggest that the success of these modes resides in the potential for service providers to capitalize on heterogenous preferences among consumers.  As such, insofar that public transit agencies and the public sector appear less well-suited to nimbly meet heterogeneous consumer demand, the private sector appears poised to play the most direct role in leveraging these new technologies towards meeting user needs.  These findings beg the question whether the public sector should play the role of actively providing services using these new technologies or of regulating downstream impacts of new mobility services and/or partnering with the private sector.

Yi Hou, National Renewable Energy Laboratory (NREL)
Venu Garikapati, National Renewable Energy Laboratory (NREL)
Dustin Weigl, National Renewable Energy Laboratory (NREL)
Alejandro Henao, National Renewable Energy Laboratory (NREL)
Matthew Moniot, National Renewable Energy Laboratory (NREL)
Joshua Sperling, National Renewable Energy Laboratory (NREL)

In the past decade, Transportation Network Companies (TNCs) such as Uber, Lyft, and Via have established themselves as a viable transportation alternative to other modes. However, the popularity of the TNC mode has come with a fair share of criticism for its negative externalities such as increasing vehicle miles traveled and congestion in cities. Ride pooling (or ridesharing), in which all or a part of an individual (or group)’s trip is shared with another individual (or group)’s trip, has the potential to reduce these externalities. Ride pooling is an effective solution to reduce congestion and travel cost, but pooled rides from TNCs still represent a small percentage of their total trips served (and miles driven), relative to single-occupancy (and without customer) vehicle miles. Both TNCs and cities alike will benefit from understanding what factors encourage or deter sharing a TNC trip. In this study, the newly available Chicago transportation network provider data was explored to identify the extent to which different socio-economic, spatio-temporal, and trip characteristics impact willingness-to-share (WTS) in ride-hailing trips. Multivariate linear regression and machine learning models were employed to understand and predict WTS based on location, time, and trip factors. The results show intuitive trends, with income level at drop-off and pick-up locations, and airport trips as the most important predictors of WTS. Results from this study can help TNCs and cities devise strategies that increase pooled ride-hailing, thereby reducing adverse transportation and energy impacts from TNC modes.