Skip to main content
ICTD 2024 Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
ViewAttendees 2

APOD Abstract for presentation (Poster or Podium)

New and Emerging Technologies

1649554 - Optimal Fleet Size Determination of a Semi-Autonomous Ride-Sourcing System under Multiple Objectives

Monday, June 17, 2024
2:30 PM - 2:45 PM ET
Room: 222
Abstract: Ride-sourcing companies such as Uber and Lyft offer riders point-to-point rides via smartphone apps. Autonomous vehicle (AV) manufacturers and ride-sourcing companies are partnering to deploy AVs in ride-sourcing systems; for example, Toyota and Daimler have partnered with Uber. Thereby, such a system becomes semi-autonomous with both human-driven vehicles (HDVs) and AVs. A typical semi-autonomous system has the following major components: (i) the operator, (ii) the drivers, (iii) the riders, (iv) HDVs, and (v) AVs (which are centrally controlled by the operator). The first three components have varying objectives (i.e., the operator wants to maximize their revenue, the drivers want to maximize their income, and the riders want to minimize their disutility). In this context, a critical question from the operator’s perspective is what should the AV fleet size be to meet demand. This is challenging to address because of the: (i) presence of HDVs in the system, (ii) conflicting objectives of the operator, drivers, and riders, and (iii) riders’ varying perceptions of AVs. This study aims to develop analytical models to determine the optimal AV fleet size. A multi-objective optimization framework is proposed that simultaneously handles conflicting objectives and maximizes the demand serviced (i.e., the number of rider-vehicle matchings). Numerical studies show that the operator can significantly increase revenue using the proposed models. These models have strong potential to aid ride-sourcing system operators in the planning of semi-autonomous systems.

Learning Objectives:

Attendees can expect to learn the following from this session:
  • Describe the conflicting objectives in a semi-autonomous ride-sourcing system.
  • Define a surge pricing multiplier that can result in the maximum number of rider-vehicle matchings.
  • Determine the optimal autonomous vehicle fleet size of a semi-autonomous ride-sourcing system under conflicting objectives.