Online Large-Scale Taxi Assignment: Optimization and Learning

We propose a solution method for online vehicle routing, which integrates machine learning routine to improve tours’ quality. Our optimization model is based on the Bertsimas et al. (2019) re-optimization approach. Two separate routines are developed. The first one uses neural network produce realistic pick-up times customers serve. second relies Q-learning in addition random walks construction of backbone graph corresponding instance problem each time step. gives improved results compared original