T-SCORE Project M2: Multi-Agent Simulation: A Model of Ride-Hailing Driver Participation

The upsurge in ride-hailing services and their rapidly growing demand have recently ignited considerable research interest. However, present research on ride-hailing is mainly focused on the demand for trips, while less attention is given to the supply enabling them, i.e., the drivers. To obtain a comprehensive understanding of ride-hailing demand, a good understanding of driver participation should be achieved. The present study aims to obtain a realistic representation of driver participation that will later be embedded in a multi-agent simulation. To date, ride-hailing research has been hampered by a lack of data, yet here the authors leverage a unique dataset of Lyft and Uber vehicle traces collected in San Francisco. Using a choice modeling approach, the authors model driver participation within four steps, estimating: (1) number of shifts on the working day, (2) shift duration, (3) shift start time, and (4) shift start location. Driver type (full-time, part-time, occasional) was found to be a strong determinant of driver participation. Full-time, but not occasional, drivers were found to work both more shifts and longer shifts. A higher number of shifts started in the downtown area (where population and employment are higher) and in higher income areas, while less shifts started in areas of high student density. Based on these modeling results, a driver fleet was generated for a typical weekday. This is one of the first studies to estimate ride-hailing driver behavior, offering insights that can potentially support transit agencies in effectively planning and regulating multi-modal transportation.

• Record URL:
  • https://rosap.ntl.bts.gov/view/dot/72993
• Supplemental Notes:
  • This document was sponsored by the U.S. Department of Transportation, University Transportation Centers Program. Cover date: January 2023.
• Corporate Authors:

  Georgia Institute of Technology, Atlanta

  School of Civil and Environmental Engineering
  790 Atlantic Drive
  Atlanta, GA  United States  30332-0355

  Transit-Serving Communities Optimally, Responsively, and Efficiently Center (T-SCORE)

  Georgia Institute of Technology
  Atlanta, Georgia  United States  30332

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Authors:
  • Tenenboim, Einat
  • 0000-0001-6018-2182
  • Xu, Yufei
  • Macfarlane, Gregory
  • 0000-0003-3999-7584
  • Erhardt, Gregory
  • 0000-0001-8133-3381
  • Peeta, Srinivas
  • 0000-0002-4146-6793
• Publication Date: 2023-3

Language

• English

Media Info

• Media Type: Digital/other
• Edition: Final Report
• Features: Appendices; Figures; References; Tables;
• Pagination: 27p

Subject/Index Terms

• TRT Terms: Behavior; Choice models; Drivers; Ridesourcing; Shifts; Simulation
• Geographic Terms: San Francisco (California)
• Subject Areas: Highways; Passenger Transportation; Planning and Forecasting; Safety and Human Factors;

Filing Info

• Accession Number: 01904547
• Record Type: Publication
• Report/Paper Numbers: TSCORE-M2
• Contract Numbers: 69A3552047141
• Files: UTC, NTL, TRIS, USDOT
• Created Date: Jan 16 2024 9:02AM