A review on reinforcement learning methods for mobility on demand systems

2025-01-05

AI Summary

This study addresses two core decision-making problems in on-demand mobility (MoD) systems: vehicle assignment and dynamic rebalancing. We propose the first unified sequential decision-making framework to systematically evaluate reinforcement learning (RL) methods—including DQN, PPO, and A3C—in realistic traffic simulation environments (SUMO/MATSim). By integrating Markov decision process modeling with multi-agent coordination mechanisms, we characterize the performance boundaries of mainstream RL algorithms under critical challenges: sparse rewards, system scalability, and the realism-deployment gap. Our analysis establishes principled mappings between algorithmic choices and operational scenario characteristics. Furthermore, we introduce a novel scheduling paradigm that balances theoretical rigor with engineering feasibility. The framework provides a reusable methodological foundation and practical guidelines for RL-driven optimization in next-generation intelligent transportation systems. (149 words)

Technology Category

• Planning, Routing, and Scheduling: Planning/Scheduling and Learning
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Mobility on Demand (MoD) refers to mobility systems that operate on the basis of immediate travel demand. Typically, such a system consists of a fleet of vehicles that can be booked by customers when needed. The operation of these services consists of two main tasks: deciding how vehicles are assigned to requests (vehicle assignment); and deciding where vehicles move (including charging stations) when they are not serving a request (rebalancing). A field of research is emerging around the design of operation strategies for MoD services, and an increasingly popular trend is the use of learning based (most often Reinforcement Learning) approaches. We review, in this work, the literature on algorithms for operation strategies of MoD systems that use approaches based on Reinforcement Learning with a focus on the types of algorithms being used. The novelty of our review stands in three aspects: First, the algorithmic details are discussed and the approaches classified in a unified framework for sequential decision-making. Second, the use cases on which approaches are tested and their features are taken into account. Finally, validation methods that can be found across the literature are discussed. The review aims at advancing the state of the art by identifying similarities and differences between approaches and highlighting current research directions.

Problem

• Reinforcement Learning
• Vehicle Allocation
• On-Demand Mobility Systems

Innovation

• Reinforcement Learning
• On-demand Mobility Services
• Algorithm Review