Scheduling of Automated Vehicles for Ridesharing Services
Shuo Han, Fei Miao, Ufuk Topcu, George Pappas

Citation
Shuo Han, Fei Miao, Ufuk Topcu, George Pappas. "Scheduling of Automated Vehicles for Ridesharing Services". Talk or presentation, October, 2016; Poster presented at the 2016 TerraSwarm Annual Meeting.

Abstract
Ridesharing services often require constant rebalancing of vehicle supplies in order to meet passenger demand across the transportation network. Current ridesharing services are mostly provided by human drivers, where rebalancing is controlled by setting price differences between service regions so that drivers are incentivized to relocate to regions with higher prices. In recent years, there has been a growing interest in introducing autonomous cars into ridesharing. Since autonomous cars can receive commands from a central authority and be directly dispatched, they provide more flexibility in terms of control and can potentially improve the efficiency of rebalancing. In our poster, we present our work on two topics in automated ridesharing. First, we proposed a metric that quantifies the efficiency gain of automated ridesharing by comparing the total amount of vehicle flow required for rebalancing. The proposed metric is independent of the actual demand and only relies on properties of the transportation network. We have also developed a set of numerical tools for computing the metric, which are applied to a case study of the Washington, DC metropolitan area. Second, we studied how to make use of predictions of future passenger demand from historical data in order to improve the service quality (measured by wait time of passengers) of automated ridesharing. We developed a data-driven distributionally robust optimization framework to address the uncertainties in demand prediction. We also showed that the robust optimization problem can be converted to an equivalent form for which numerically efficient solutions are available. We evaluate the performance of the data-driven vehicle balancing framework based on four years of taxi trip data from New York City.

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  Shuo Han, Fei Miao, Ufuk Topcu, George Pappas. <a
  href="http://www.terraswarm.org/pubs/860.html"><i>Scheduling
  of Automated Vehicles for Ridesharing
  Services</i></a>, Talk or presentation, 
  October, 2016; Poster presented at the <a
  href="http://terraswarm.org/conferences/16/annual"
  >2016 TerraSwarm Annual Meeting</a>.

• Plain text

  Shuo Han, Fei Miao, Ufuk Topcu, George Pappas.
  "Scheduling of Automated Vehicles for Ridesharing
  Services". Talk or presentation,  October, 2016; Poster
  presented at the <a
  href="http://terraswarm.org/conferences/16/annual"
  >2016 TerraSwarm Annual Meeting</a>.

• BibTeX

  @presentation{HanMiaoTopcuPappas16_SchedulingOfAutomatedVehiclesForRidesharingServices,
      author = {Shuo Han and Fei Miao and Ufuk Topcu and George
                Pappas},
      title = {Scheduling of Automated Vehicles for Ridesharing
                Services},
      month = {October},
      year = {2016},
      note = {Poster presented at the <a
                href="http://terraswarm.org/conferences/16/annual"
                >2016 TerraSwarm Annual Meeting</a>.},
      abstract = {Ridesharing services often require constant
                rebalancing of vehicle supplies in order to meet
                passenger demand across the transportation
                network. Current ridesharing services are mostly
                provided by human drivers, where rebalancing is
                controlled by setting price differences between
                service regions so that drivers are incentivized
                to relocate to regions with higher prices. In
                recent years, there has been a growing interest in
                introducing autonomous cars into ridesharing.
                Since autonomous cars can receive commands from a
                central authority and be directly dispatched, they
                provide more flexibility in terms of control and
                can potentially improve the efficiency of
                rebalancing. In our poster, we present our work on
                two topics in automated ridesharing. First, we
                proposed a metric that quantifies the efficiency
                gain of automated ridesharing by comparing the
                total amount of vehicle flow required for
                rebalancing. The proposed metric is independent of
                the actual demand and only relies on properties of
                the transportation network. We have also developed
                a set of numerical tools for computing the metric,
                which are applied to a case study of the
                Washington, DC metropolitan area. Second, we
                studied how to make use of predictions of future
                passenger demand from historical data in order to
                improve the service quality (measured by wait time
                of passengers) of automated ridesharing. We
                developed a data-driven distributionally robust
                optimization framework to address the
                uncertainties in demand prediction. We also showed
                that the robust optimization problem can be
                converted to an equivalent form for which
                numerically efficient solutions are available. We
                evaluate the performance of the data-driven
                vehicle balancing framework based on four years of
                taxi trip data from New York City.},
      URL = {http://terraswarm.org/pubs/860.html}
  }
  

Posted by Shuo Han on 26 Oct 2016.
Groups: services