Analysis of Transportation Network Company Survey Data, App-Based Ridesharing Data and Previous Literature Provides Insights on How Differing User Preferences Can Be Accommodated to Increase the Proportion of Shared Trips.
Nationwide, United States
Analysis of Travel Choices and Scenarios for Sharing Rides
Summary Information
Increasing automobile occupancy by sharing rides can help reduce overall vehicle miles travelled (VMT), and thereby alleviate congestion, curtail vehicle emissions and improve the efficiency of highway operations. This study aimed to describe the effect of price and time on a rider’s choice between private party and shared ride hailing. The study sought to understand behavior and the impacts of price and time tradeoffs for two models of sharing private rides, including 1) Shared Options in ridehailing applications (mobility apps that allow users to select private rides and provide an upfront estimate on price and time), and 2) Ridesharing using app-based incentive tools (emerging apps that allow traditional carpooling matches to be created in real time). Impacts of four changes in modal travel characteristics that could relate to potential policy mechanisms were evaluated:
- Increased cost savings for shared transportation network company (TNC) trips relative to private TNC trips.
- Reduced travel time penalty for shared TNC trips relative to private TNC trips.
- Increased price differential of private car trips relative to all other modes.
- Rewarding carpooling trips but not private car trips.
This study used data on revealed preferences for private party and shared ridehailing trips in 15 American cities, coupled with a large Transportation Network Company's (TNC) survey of 4,365 of its users that was administered in late 2018.
- Offer customers more than one shared product option with time delay and price points to increase the proportion of shared trips. A high degree of heterogeneity in user preferences was found in the study. This suggests that more options should be provided in combination with vehicle routing decisions designed to accommodate differing user preferences.
- Provide targeted incentives by focusing on particular geographies or population segments to reduce Vehicle Miles Travelled (VMT). For example, the effect of increasing price differences was greater for riders with annual income under $50,000, for trips that begin in dense office districts, and trips that end in areas with competitive transit. Trips to or from the airport or other intermodal travel centers and trips paid for by a third party were very insensitive to price, likely due to time sensitivity.
- Reduce travel time penalty via changes in waiting time for shared TNC trips relative to private TNC trips to reduce VMT. Considering all trips in all study cities, a 15 seconds per mile travel time difference reduction between private and shared TNCs would reduce VMT by roughly 85 million miles per year by reducing private TNC miles by 11.9 percent and substituting shared TNC travel for that difference. This differential can be difficult to obtain from travel time changes and could more easily result from changes in waiting time (e.g., prioritizing pick-ups for shared TNC rides).
- Understand that price and time are not the only influential factors on shared rides. Based on the survey results, over 30 percent of users rejected a shared trip that cost 75 percent less than the observed private trip, even when the presented travel time was identical. This reflects the fact that unwillingness to share was not only related to price and time.
- Prioritize strategies on increasing the price differential of private car trips relative to all other modes. Lower relative prices and faster relative travel time for shared TNC trips have a much smaller impact on VMT because TNCs currently represent a very small portion of total trips and total VMT (about 1 percent of all person trips in the study city regions). Considering all trips in all study city regions, a $1 per trip relative price increase for private trips would reduce VMT by roughly 3.6 billion miles per year.
