发表机构
Universidade da Coruña（科鲁尼亚大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究评估通用LLM作为智能体，无需任务特定训练即可从自然语言描述制定并实现约束规划调度模型，多智能体架构显著提升实现成功率，但紧密耦合物流模型仍具挑战。
AI 中文摘要
开发生产调度的优化模型需要大量的专家努力。关于大型语言模型（LLM）的研究遵循了两个方向：专门用于自动化建模的方法，主要针对混合整数线性规划，这些方法通常依赖于专用训练或特定问题的架构，限制了工业部署；以及用于操作决策支持的智能体人工智能，这通常假设优化模型已经存在。本研究通过评估通用LLM（作为智能体编排，无需任务特定训练）能否从自然语言问题描述中制定并实现约束规划模型，来弥合这两个方向。单智能体和多智能体架构与一个模型上下文协议服务器集成，该服务器提供对求解器文档的上下文感知检索，以减轻实现过程中的幻觉。两者都与直接LLM基线在六个面向工业的问题上进行了比较，涵盖流水车间、作业车间、柔性作业车间和资源受限的仓库调度，使用三种LLM并评估建模准确性、执行成功率、延迟和令牌消耗。公式化证明在很大程度上已处于当前LLM的能力范围内，而实现是主要障碍。多智能体工作流将从直接LLM调用生成的正确运行脚本比例从14.8%提高到59.3%，在四个较不复杂的问题上达到80.6%，而紧密耦合的物流内部模型仍然是一个开放的挑战。
英文摘要
Developing optimization models for production scheduling requires substantial expert effort. Research on large language models (LLMs) has followed two directions: specialized approaches for automated modeling, mostly for mixed-integer linear programming, which often rely on dedicated training or problem-specific architectures that limit industrial deployment; and agentic artificial intelligence for operational decision support, which generally assumes that the optimization model already exists. This study bridges both directions by assessing whether general-purpose LLMs, orchestrated as agents without task-specific training, can formulate and implement constraint programming models from natural-language problem descriptions. Singleagent and multi-agent architectures are integrated with a Model Context Protocol server that provides context-aware retrieval of solver documentation to mitigate hallucinations during implementation. Both are compared with a direct LLM baseline on six industry-oriented problems covering flow-shop, job-shop, flexible job-shop and resource-constrained warehouse scheduling, using three LLMs and assessing modeling accuracy, execution success, latency and token consumption. Formulation proves largely within reach of current LLMs, whereas implementation is the main barrier. The multi-agent workflow raises the share of scripts that run correctly as generated from 14.8% with a direct LLM call to 59.3%, reaching 80.6% on the four less complex problems, while tightly coupled intralogistics models remain an open challenge.
Comments29 pages, 9 figures
