The Value of Knowing Drivers’ Opportunity Cost in Ride Hailing Systems
Video
Team Information
Team Members
Ran I. Snitkovsky, Researcher, Columbia Business School and Shenzhen Research Institute of Big Data (SRIBD)
Costis Maglaras, David and Lyn Silfen Professor of Business, Columbia Business School
Jim G. Dai, Leon C. Welch Professor of Engineering, School of Operations Research and Information Engineering, Cornell University
Abstract
Consider a ride hailing platform, and a large population of strategic potential drivers, heterogeneous in terms of their opportunity costs, who choose whether or not to work for that platform. The platform is provided with knowledge about the different drivers' opportunity costs. How can the platform implement a matching policy that uses this knowledge in order to improve system efficiency? Can such improvement be quantified? In this work we introduce a mean field (fluid) model that accounts for the dynamic nature of drivers' spatial location, revenue and availability, allowing us to analyze drivers' strategic behavior. Our analysis leads to improvement bounds on the equilibrium performance: We show that a policy which utilizes knowledge about drivers' opportunity costs can perform up to 2 times better than a similar policy that ignores it, in terms of the number of drivers it attracts and in terms of the rate of matches it produces. We demonstrate by simulation that the mean field model provides an accurate approximation for a corresponding discrete model, in which such improvements are empirically demonstrated.
Contact this Team
Team Contact: Ran I. Snitkovsky (use form to send email)