LLM-ODDR: A Large Language Model Framework for Joint Order Dispatching and Driver Repositioning

TL;DR

This work introduces LLM-ODDR, a large language model–based framework for jointly optimizing ride-hailing order dispatching and driver repositioning. It fuses three components—Multi-objective-guided Order Value Refinement, Fairness-aware Order Dispatching, and Spatiotemporal Demand-Aware Driver Repositioning—into a cohesive decision system powered by a fine-tuned JointDR-GPT (LoRA-enhanced) trained on domain data. Empirical evaluation on Manhattan taxi datasets demonstrates notable gains in GMV and Order Response Rate, especially under surge conditions, along with improved interpretability and fairness in driver earnings. While the approach yields strong performance and adaptability, it incurs higher computation costs, motivating future work on latency reduction and more scalable inference techniques.

Abstract

Ride-hailing platforms face significant challenges in optimizing order dispatching and driver repositioning operations in dynamic urban environments. Traditional approaches based on combinatorial optimization, rule-based heuristics, and reinforcement learning often overlook driver income fairness, interpretability, and adaptability to real-world dynamics. To address these gaps, we propose LLM-ODDR, a novel framework leveraging Large Language Models (LLMs) for joint Order Dispatching and Driver Repositioning (ODDR) in ride-hailing services. LLM-ODDR framework comprises three key components: (1) Multi-objective-guided Order Value Refinement, which evaluates orders by considering multiple objectives to determine their overall value; (2) Fairness-aware Order Dispatching, which balances platform revenue with driver income fairness; and (3) Spatiotemporal Demand-Aware Driver Repositioning, which optimizes idle vehicle placement based on historical patterns and projected supply. We also develop JointDR-GPT, a fine-tuned model optimized for ODDR tasks with domain knowledge. Extensive experiments on real-world datasets from Manhattan taxi operations demonstrate that our framework significantly outperforms traditional methods in terms of effectiveness, adaptability to anomalous conditions, and decision interpretability. To our knowledge, this is the first exploration of LLMs as decision-making agents in ride-hailing ODDR tasks, establishing foundational insights for integrating advanced language models within intelligent transportation systems.

Figures (6)

• Figure 1: The pipeline of large language models for joint order dispatching and driver repositioning tasks.
• Figure 2: The proposed LLM-ODDR framework for joint order dispatching and driver repositioning tasks.
• Figure 3: Fairness study of driver income for the proposed JointDR-GPT and other LLM-based models on the small-100 dataset.
• Figure 4: Comparison of order request volumes between original and synthesized datasets.
• Figure 5: Performance comparison of dispatching models during surge events.