[3I2-GS-5e] エージェント：応用

[3I2-GS-5e-05] Introducing Fairness to ride-hailing service

〇Masato Ota¹, Yuko Sakurai², Itsuki Noda² (1. Kwansei Gakuin University, 2. National Institute of Advanced Industrial Science and Technology)

Keywords:Taxi-dispatch algorithm, Fairness, Reinforcement Learning

We propose a new fair-assignment algorithm between vehicles and passengers to improve the efficiency and fairness tradeoff for on-demand ride-hailing platforms. The main feature of the proposed assignment algorithm is that it gives priority to drivers with low cumulative profit and then assigns the most preferred passenger to the selected drivers. We experimentally demonstrate that it works better than the existing algorithms in both terms of efficiency and fairness.