Optimizing Large-Scale Fleet Management on a Road Network using Multi-Agent Deep Reinforcement Learning with Graph Neural Network
This addresses the challenge of dynamic resource allocation in ride-hailing for improved efficiency, though it is incremental as it builds on existing reinforcement learning methods with a graph-based spatial representation.
The paper tackles the problem of optimizing large-scale fleet management for ride-hailing services by representing the road network as a graph and using multi-agent deep reinforcement learning with graph neural networks, achieving superior results over greedy policy updates in simulations based on empirical taxi data.
We propose a novel approach to optimize fleet management by combining multi-agent reinforcement learning with graph neural network. To provide ride-hailing service, one needs to optimize dynamic resources and demands over spatial domain. While the spatial structure was previously approximated with a regular grid, our approach represents the road network with a graph, which better reflects the underlying geometric structure. Dynamic resource allocation is formulated as multi-agent reinforcement learning, whose action-value function (Q function) is approximated with graph neural networks. We use stochastic policy update rule over the graph with deep Q-networks (DQN), and achieve superior results over the greedy policy update. We design a realistic simulator that emulates the empirical taxi call data, and confirm the effectiveness of the proposed model under various conditions.