The code contained in this repository represents a Python implementation of REASSIGN Algorithm depicted in:
Balancing Efficiency and Fairness in On-Demand Ridesourcing (in Proc. NeurIPS 2019)
Nixie S. Lesmana, Xuan Zhang, Xiaohui Bei
REASSIGN is a graph-based reassignment algorithm that balances two objectives in ridesourcing platform decision making: system efficiency and driver fairness. It allows flexible shifting between decisions with trade-off guarantees.
The repository is organized into folders and subfolders, which has been named according to the scripts directory that we used in the codes. A few things to note:
- Main scripts for our experiments can be found in Simulation Codes.
- In Data subfolders, several files have been omitted due to upload size limit. These files can be downloaded through this external link.
- Results do not contain our actual simulation results. These files are created as samples.
For more info: Data Preprocessing, Simulation Setup, and Code Summary