Large-scale simulation-based evaluation of fleet repositioning strategies for dynamic rideshare in New York City

Jae Young Jung, Joseph Chow

Research output: Contribution to journalConference article

Abstract

There has been a growing concern about increasing vehicle-mile traveled (VMT) associated with deadhead trips for dynamic rideshare services, particularly with the emergence of Shared Autonomous Vehicle (SAV) services. Studies in the literature on repositioning strategies have been limited to synthetic or small-scale study areas. This study considers a large-scale computational experiment involving a New York City study area with a network of 16,782 nodes and 23,337 links with 662,455 potential travelers from the 2016 Yellow Taxi data. We investigate the potential to reduce VMT and deadhead miles for dynamic rideshare operations combined with vehicle repositioning strategies. Three repositioning strategies are evaluated: (1) Roaming around areas with higher pickup probabilities to maximize the chance of picking up passengers, (2) Staying at curb side after completing trips, and (3) Repositioning to depots to minimize deadhead trips. The study suggests the last strategy of having optimized depots can minimize both trip rejections and passenger journey time at the expense of increased VMT, although the amount depends significantly on fleet size.

Original languageEnglish (US)
JournalSAE Technical Papers
Volume2019-April
Issue numberApril
DOIs
StatePublished - Apr 2 2019
EventSAE World Congress Experience, WCX 2019 - Detroit, United States
Duration: Apr 9 2019Apr 11 2019

    Fingerprint

ASJC Scopus subject areas

  • Automotive Engineering
  • Safety, Risk, Reliability and Quality
  • Pollution
  • Industrial and Manufacturing Engineering

Cite this