INFORMS Open Forum

Expand all | Collapse all

INFORMS Boston Chapter meeting and talk on Thursday, May 2nd, at 6:30 PM, talk by Hongyao Ma, "Spatio-Temporal Pricing for Ridesharing Platforms"

  • 1.  INFORMS Boston Chapter meeting and talk on Thursday, May 2nd, at 6:30 PM, talk by Hongyao Ma, "Spatio-Temporal Pricing for Ridesharing Platforms"

    Posted 04-17-2019 20:20

    Please join us for an INFORMS Boston Chapter meeting and talk by Hongyao Ma, "Spatio-Temporal Pricing for Ridesharing Platforms"

    The event will be held at Bentley University and is co-sponsored by Bentley's Data Analytics Research Team (DART)

    Time: Thursday May 2nd, 2019. 6:30 pm to 8:00 pm. Light refreshments will be served at 6:30.

    Location:  Center for Marketing Technology (CMT), Morison 200, Bentley University

    Title: Spatio-Temporal Pricing for Ridesharing Platforms

    Speaker: Hongyao Ma, a Ph.D. candidate in Computer Science in Harvard SEAS, will join Uber from summer 2019.

    Abstract:

    As we envision a future with an ever-increasing integration of technology and markets, the questions of reliability come to the core. In this talk, I present my work on matching and pricing for ridesharing platforms, where drivers' strategic behavior (e.g. cherry-picking trips, and declining trips to chase surge prices) undercut their stated mission of "providing transportation as reliable as running water." I introduce the Spatio-Temporal Pricing mechanism, which uses information about supply and demand over a planning horizon, solves for the welfare-optimal matching, and sets prices that are "smooth" in both space and time. In this way, the mechanism aligns incentives for drivers and promotes reliability without using penalties or time-extended contracts--- always accepting all trip dispatches forms a subgame-perfect equilibrium among the drivers. The mechanism also satisfies fairness properties, including that drivers at the same location and time do not envy each other's continuation payoffs. The mechanism is also robust, in that it achieves optimality along with other axiomatic properties from any history onward, including after a deviation. Simulation analysis suggests that the STP mechanism can achieve significantly higher social welfare than a myopic pricing mechanism.

    Bio:

    Hongyao Ma is a Ph.D. candidate in Computer Science in Harvard SEAS, advised by Prof. David C. Parkes. Her research is broadly situated at the intersection of economics and computer science, and she is particularly interested in designing market-based systems to bring people together in useful ways, in the presence of uncertainty, self-interest, and autonomy. Ma is a recipient of the Siebel Scholarship, Harvard SEAS fellowship, and the Certificate of Distinction in Teaching at Harvard. She received her M.S. in 2014 at Harvard, and B.E. in 2012 at Xi'an Jiaotong University, both in Electrical Engineering.  



    ------------------------------
    Daniel Rice
    Senior Principal Scientist
    FAST Labs™, BAE Systems
    ------------------------------


  • 2.  RE: INFORMS Boston Chapter meeting and talk on Thursday, May 2nd, at 6:30 PM, talk by Hongyao Ma, "Spatio-Temporal Pricing for Ridesharing Platforms"

    This message was posted by a user wishing to remain anonymous
    Posted 04-18-2019 07:13
    This post was removed