A Reinforcement Learning and Prediction-Based Lookahead Policy for Vehicle Repositioning in Online Ride-Hailing Systems

Existing approaches for vehicle repositioning on large-scale ride-hailing platforms either ignore the spatial-temporal mismatch between supply and demand in real-time or overlook the long-term balance of the system. To account for both, the authors propose a lookahead repositioning policy in this paper, which is a novel approach to repositioning idle vehicles from both a dynamic system and a long-term performance perspective. Their method consists of two parts; the first part utilizes linear programming (LP) to formulate the nonstationary system as a time-varying, 𝑇-step lookahead optimization problem and explicitly models the fraction of drivers who follow repositioning recommendations (called the repositioning rate). The second step is to incorporate a reinforcement learning (RL) method to maximize long-term return based on learned value functions after the 𝑇 time slots. Extensive studies utilizing a real-world dataset on both small-scale and large-scale simulators show that their method outperforms previous baseline methods and is robust to prediction errors.

• Record URL:
  • https://doi.org/10.1109/TITS.2023.3312048
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/41297384
• Supplemental Notes:
  • Copyright © 2024, IEEE.
• Authors:
  • Wei, Honghao
  • Yang, Zixian
  • Liu, Xin
  • Qin, Zhiwei
  • Tang, Xiaocheng
  • Ying, Lei
• Publication Date: 2024-2

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: pp 1846-1856
• Serial:
  • IEEE Transactions on Intelligent Transportation Systems
  • Volume: 25
  • Issue Number: 2
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 1524-9050
  • Serial URL: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6979

Subject/Index Terms

• TRT Terms: Dynamic positioning; Fleet management; Machine learning; Predictive models; Ridesourcing
• Subject Areas: Data and Information Technology; Highways; Planning and Forecasting;

Filing Info

• Accession Number: 01923091
• Record Type: Publication
• Files: TRIS
• Created Date: Jun 27 2024 2:11PM