Short-term repositioning for empty vehicles on ride-sourcing platforms

• Main Authors: WANG, Hai, WANG, Zhengli
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2020
• Subjects: Artificial Intelligence and Robotics
• Online Access: https://ink.library.smu.edu.sg/sis_research/6113; https://ink.library.smu.edu.sg/context/sis_research/article/7116/viewcontent/39_Hai_Wang_Optimal_Dispatch_TSL_Extended_Abstract__1_.pdf
• Institution: Singapore Management University

Motivation Ride sourcing companies, such as Uber, Lyft, and Didi, have been able to leverage on internet-based platforms to connect passengers and drivers. These platforms facilitate passengers and drivers’ mobility data on smartphones in real time, which enables a convenient matching between demand and supply. The imbalance of demand (i.e., passenger requests) and supply (i.e., drivers) on the platforms causes many unserved passenger requests and empty vehicles with idle drivers to exist at the same time, which poses a challenging problem for the platform. To address these challenges, some platforms display heat maps of surge-pricing multipliers or real-time demand to drivers, and anticipate that such information will induce more idle drivers to self-reposition by cruising to regions with high demand and/or low supply. Some platforms attempt to provide direct repositoning guidance to drivers, by suggesting that drivers cruise to a specific region. In practical terms, drivers are more likely to follow guidance that directs them to a region close to their current location.