PortAgent: LLM-driven Vehicle Dispatching Agent for Port Terminals
🤖 AI Summary
Automated Container Terminal (ACT) Vehicle Dispatch Systems (VDS) face three critical bottlenecks in cross-terminal deployment: heavy reliance on domain experts, substantial local data requirements, and prolonged deployment cycles. To address these, this paper proposes an LLM-driven vehicle dispatch agent featuring a novel Virtual Expert Team (VET) architecture—enabling zero expert dependency, minimal data requirements, and minute-scale deployment. We introduce the first integration of Retrieval-Augmented Generation (RAG) with Reflexion-based self-refinement into a closed-loop LLM reasoning paradigm, supporting fully automated VDS migration across terminals. Leveraging few-shot domain adaptation, collaborative virtual expert reasoning, and automated code generation and debugging, our system reduces cross-terminal deployment time by 90% and generates executable dispatch policies from only 3–5 demonstration examples. This significantly enhances VDS portability and engineering deployability.
Application Category
📝 Abstract
Vehicle Dispatching Systems (VDSs) are critical to the operational efficiency of Automated Container Terminals (ACTs). However, their widespread commercialization is hindered due to their low transferability across diverse terminals. This transferability challenge stems from three limitations: high reliance on port operational specialists, a high demand for terminal-specific data, and time-consuming manual deployment processes. Leveraging the emergence of Large Language Models (LLMs), this paper proposes PortAgent, an LLM-driven vehicle dispatching agent that fully automates the VDS transferring workflow. It bears three features: (1) no need for port operations specialists; (2) low need of data; and (3) fast deployment. Specifically, specialist dependency is eliminated by the Virtual Expert Team (VET). The VET collaborates with four virtual experts, including a Knowledge Retriever, Modeler, Coder, and Debugger, to emulate a human expert team for the VDS transferring workflow. These experts specialize in the domain of terminal VDS via a few-shot example learning approach. Through this approach, the experts are able to learn VDS-domain knowledge from a few VDS examples. These examples are retrieved via a Retrieval-Augmented Generation (RAG) mechanism, mitigating the high demand for terminal-specific data. Furthermore, an automatic VDS design workflow is established among these experts to avoid extra manual interventions. In this workflow, a self-correction loop inspired by the LLM Reflexion framework is created
Problem
Research questions and friction points this paper is trying to address.
Automates vehicle dispatching system transfer across terminals
Reduces reliance on port specialists and terminal-specific data
Enables fast deployment with automated design workflow
Innovation
Methods, ideas, or system contributions that make the work stand out.
LLM-driven agent automates vehicle dispatching system transfer
Virtual Expert Team eliminates need for port operations specialists
Retrieval-Augmented Generation reduces terminal-specific data requirement
🔎 Similar Papers
No similar papers found.
