Abstract
Ride-sharing platforms face a “productivity paradox,” whereby any efficiency gained through improved dispatch or pricing strategies will not benefit drivers or riders. We show that this is a limit of the traditional ride-hailing model and a consequence of the Hall-Horton driver equilibrium earning hypothesis. In response to this challenge, Lyft introduced Priority Mode (PM), which allows drivers to concentrate their work during specific prioritized hours. We prove that PM solves the productivity paradox. As a result, the average driver earnings increase, and the platform and the riders also benefit. Implementing PM requires significant changes to the platform's dispatch and pricing policy but most importantly requires careful control of the number of drivers that can be offered the opportunity to be prioritized at any given time. In this paper, we introduce a queuing setting to model the market dynamics of PM and illustrate the challenges of this control problem. We then leverage this intuition to build a real-time priority admission control system that can balance the number of drivers offered priority and achieve the desired productivity increase. Lyft has successfully rolled out PM throughout North America, and drivers have completed hundreds of thousands of driving hours thus far. It has generated tens of millions of dollars of value that the drivers, the riders, and Lyft have shared, with the potential to generate much more when rolled out in all markets. Finally, our internal driver surveys reveal that it has been well received by drivers.
Original language | English (US) |
---|---|
Pages (from-to) | 433-445 |
Number of pages | 13 |
Journal | INFORMS Journal on Applied Analytics |
Volume | 52 |
Issue number | 5 |
DOIs | |
State | Published - Sep 2022 |
Funding
The authors thank Hao Yi Ong, Omkar Savant, Barrett Atwood, Antonio Raimundo, Eric Wood, Tolu Adeofe, Kandra Ishihara, and Carl Bowers for generous feedback and approval throughout this process. Priority mode is still a work-in-progress, and it was a collaborative effort throughout the company. In partiular, those who helped develop the core of the algorithms described in this paper include Praveen Athmanathan, Cerek Hillen, Cody Wang, Nathan Fraenkel, Owen Tong, Di Wu, Charlene Zhou, Davide Crapis, Wilson Turner, Derek Salama, Xabi Azagirre, John Fremlin, Shannon Mcnish, Baptiste Richard, Allen Yuan, Huiling Zhang, Shashi Sharma, Patrick Lung, Vijay Narasiman, Ariel Goldszmidt, Adi Rathnam, Luke Gong, Tim Wang, Rakesht Kumar, Brianna Horton, Peter Kong, Shane Auerbach, Alex Zamoschin, Bo Zhang, and more.
Keywords
- mechanism design
- optimization
- ride-sharing
- stochastic control
ASJC Scopus subject areas
- Management of Technology and Innovation
- Strategy and Management
- Management Science and Operations Research