[2603.00176] Bridging Policy and Real-World Dynamics: LLM-Augmented Rebalancing for Shared Micromobility Systems

Computer Science > Machine Learning arXiv:2603.00176 (cs) [Submitted on 26 Feb 2026] Title:Bridging Policy and Real-World Dynamics: LLM-Augmented Rebalancing for Shared Micromobility Systems Authors:Heng Tan, Hua Yan, Yu Yang View a PDF of the paper titled Bridging Policy and Real-World Dynamics: LLM-Augmented Rebalancing for Shared Micromobility Systems, by Heng Tan and 2 other authors View PDF HTML (experimental) Abstract:Shared micromobility services such as e-scooters and bikes have become an integral part of urban transportation, yet their efficiency critically depends on effective vehicle rebalancing. Existing methods either optimize for average demand patterns or employ robust optimization and reinforcement learning to handle predefined uncertainties. However, these approaches overlook emergent events (e.g., demand surges, vehicle outages, regulatory interventions) or sacrifice performance in normal conditions. We introduce AMPLIFY, an LLM-augmented policy adaptation framework for shared micromobility rebalancing. The framework combines a baseline rebalancing module with an LLM-based adaptation module that adjusts strategies in real time under emergent scenarios. The adaptation module ingests system context, demand predictions, and baseline strategies, and refines adjustments through self-reflection. Evaluations on real-world e-scooter data from Chicago show that our approach improves demand satisfaction and system revenue compared to baseline policies, highlighting...