Analysis of Travel Choices and Scenarios for Sharing Rides
Final Report

Printable version [PDF 2.5 MB]
Contact Information: Operations Feedback at OperationsFeedback@dot.gov

U.S. Department of Transportation
Federal Highway Administration
Office of Operations
1200 New Jersey Avenue, SE
Washington, DC 20590
ops.fhwa.dot.gov

March 2021
FHWA-HOP-21-011

Table of Contents

[ Notice, Quality Assurance, and Non-Binding Contents ] [ Technical Report Documentation Page ] [ List of Abbreviations ]

• Executive Summary
• Chapter 1. Project Overview
  • Project Purpose
  • Approach and Methodology
    • Survey of Transportation Network Company Users
    • Analysis of App-Based Carpooling Use
    • Literature Review on Driving/Parking Price and Behavioral Response
    • Development of an Analytical Model for City-Level Scenario Assessments
  • Document Organization
• Chapter 2. Sharing Rides in an Environment of On-Demand Ridehailing
  • Overall Context
    • Ridehailing Literature
  • Survey Approach and Data Preparation
    • Trip Alternative Questions
    • Appended Data on Observed Trips
    • Sample Cleaning
    • Data Weighting
    • Discrete Choice Model Development
    • Passenger Occupancy in Private Rides
    • Limitations of Approach
  • Analysis of Market Segments
    • Frequency of Sharing by User Characteristic
    • Distribution of Shared and Private Trips by Trip Characteristic
    • Distribution of Shared and Private Trips by Built Environment Characteristic
  • Analysis of Price and Travel Time Effects on Choice to Share
    • Effect of Price on Sharing
    • Market Segmentations and Price Sensitivity
    • Effect of Time on Sharing
    • Relative Effect of Price and Time
  • Discussion: Ridesharing and Transportation Network Companies
• Chapter 3. Encouraging Carpooling Using App-Based Incentive Tools
  • Impact of Rewards and Incentives using Data from Metropia™ Driving Up Occupancy Platform
  • Impact of Rewards and Incentives using Examples from Hytch™
• Chapter 4. Shared Ride Use Under Different Scenarios
  • Scenario Details
    • Overview of Scenarios
    • Use of the Scenarios in the Analytic Model
  • Analytic Model For Scenario Assessments
    • Analytic Model Inputs and Assumptions
    • Interpreting Analytic Model Outputs
    • Limitations of the Analytic Model
  • Scenario Analysis Results
    • Scenario 1: Lower Relative Prices for Shared Transportation Network Company Trips
    • Scenario 2: Faster Relative Travel Time for Shared Transportation Network Company Trips
    • Combination of Scenarios 1 and 2: Greater Price Differential and Less Time Differential
    • Scenario 3: Increase Price Difference of Private Car Trips and All Other Modes
• Chapter 5. Implications for Cities
• Chapter 6. Potential for Future Research
• References
• Acknowledgements

List of Figures

• Figure 1. Graph. Distribution of observed frequency of sharing according to user characteristic. Bars of the same frequency level sum to 100 in each panel (n = 4,365). (Source: Federal Highway Administration)
• Figure 2. Graph. Distribution of shared and private trips according to trip characteristics (n = 4,365). (Source: Federal Highway Administration)
• Figure 3. Graph. Distribution of shared and private trips according to the built environment characteristics. Categories on the x-axis represent percentiles (n = 4,365). (Source: Federal Highway Administration)
• Figure 4. Graph. Share of private transportation network company users that switched to shared travel. Users in dataset that switched from private to shared travel at three levels of price differences offered (with average price differential shown in parentheses). These three alternatives represented shared trips with identical travel times to the observed private trip. (Source: Federal Highway Administration)
• Figure 5. Graph. Effect of $1 per mile relative price difference on a user's percent probability of sharing rides. (Source: Federal Highway Administration)
• Figure 6. Graph. Share of private transportation network company users in dataset that switched from private to shared travel at each level of travel time difference and price difference offered. (Source: Federal Highway Administration)
• Figure 7. Graph. Effect of 1 minute/mile reduction in relative travel time difference on the percent probability of sharing. (Source: Federal Highway Administration)
• Figure 8. Graph. Driving Up Occupancy timeframes and user data definition. (Source: Metropia™)
• Figure 9. Diagram. Tiered analysis framework. (Source: Metropia™)
• Figure 10. Graph. Temporal carpool use. (Source: Metropia™)
• Figure 11. Graph. Carpool driver and passenger incentive point trend. (Source: Metropia™)
• Figure 12. Diagram. Total miles per user per month compared to average rewards per mile. (Source: Middle Tennessee State University Data Science Institute)
• Figure 13. Percentage of trips with no rewards and last trips per month. (Source: Middle Tennessee State University Data Science Institute)
• Figure 14. Graphic. Explanation of modes and decisions affected by scenarios 1 and 2. (Source: Federal Highway Administration)
• Figure 15. Graphic. Explanation of modes and decisions affected by scenario 3. (Source: Federal Highway Administration)
• Figure 16. Graphic. Explanation of modes and decisions affected by experimental scenario 4. (Source: Federal Highway Administration)
• Figure 17. Image. Screenshots, from left to right: "Scenarios" sidebar tab, "Segments" sidebar tab, and "Customize Initial Mode Shares" sidebar tab. (Source: Federal Highway Administration)
• Figure 18. Image. Screenshot of slider bars features in "Assumptions" sidebar tab. (Source: Federal Highway Administration)
• Figure 19. Graph. Sample output of analytic model for illustrative purposes. (Source: Federal Highway Administration)
• Figure 20. Graph. Effect of scenarios on vehicle miles traveled (total change and percent change). (Source: Federal Highway Administration)

List of Tables

• Table 1. Summary of findings
• Table 2. Alternative options examples. (Source: Federal Highway Administration)
• Table 3. Characteristics of cards presented to private and shared transportation network company users in trip alternative analysis. (Source: Federal Highway Administration)
• Table 4. Share of adjusted survey data at three occupancy levels for private ridehailing trips (n = 3,518). (Source: Federal Highway Administration)
• Table 5. Exponentiated coefficients of Model 8. (Source: Federal Highway Administration)
• Table 6. Initial rate of opting in to shared rides for selected market segments. (Source: Federal Highway Administration)
• Table 7. Reported reasons why respondents chose a private ride over a shared ride (values do not add to 100 percent because respondents could select more than one reason). (Source: Federal Highway Administration)
• Table 8. Illustration of effect of price and time differences on overall level of sharing. (Source: Federal Highway Administration)
• Table 9. Implied value of time based on choice between 11 pairs of shared ride options. (Source: Federal Highway Administration)
• Table 10. Implied value of time based on choice between nine pairs of shared ride options. (Source: Federal Highway Administration)
• Table 11. Number of Metropia™ users contacted via the micro-survey, by market. (Source: Metropia™)
• Table 12. Distribution of carpool users included in the trend analysis. (Source: Metropia™)
• Table 13. Counts and percentages for no reward and last trip (Source: Middle Tennessee State University Data Science Institute)
• Table 14. Average rewards per mile per month for a user (Source: Middle Tennessee State University Data Science Institute)
• Table 15. Weekly totals for a Hytch™ rewards partner (Source: Middle Tennessee State University Data Science Institute)
• Table 16. Overview of scenarios considered for analysis.
• Table 17. Average cost of drive-alone trips in study cities, with and without parking costs. (Sources: City Observatory Price of Parking, Shoup (2005), National Household Travel Survey (2017))
• Table 18. Sample effect of increasing relative per-trip drive-alone costs on drive-alone mode share according to Trip Reduction Impacts of Mobility Management Strategies elasticity estimate of -0.30. (Source: Federal Highway Administration)
• Table 19. Effect of $1/mile price difference for shared transportation network company trips across three geographies, for all segments, and for trips starting in dense office districts. See scenario 1. (Source: Federal Highway Administration)
• Table 20. Effect of 15 seconds/mile reduced travel time difference between private and shared transportation network company trips across three geographies, for all segments, and for trips ending in dense office districts. See scenario 2 (affects only shared transportation network company modes). (Source: Federal Highway Administration)
• Table 21. Effect of both a $1/mile increased price difference and a 15 seconds/mile reduced travel time difference between private and shared transportation network company trips across three geographies, in all segments (left) and trips ending in dense office districts (right). See scenarios 1 and 2 (affects only transportation network company modes). (Source: Federal Highway Administration)
• Table 22. Effect of $1/trip relative increase in the price of private car trips compared to all other modes (affects all modes), in all segments for three different geographies. See scenario 3. (Source: Federal Highway Administration)