Automated Constraint Specification for Job Scheduling by Regulating Generative Model with Domain-Specific Representation
School of Advanced Manufacturing and Robotics, Peking University(先进制造与机器人学院,北京大学);Department of Computer Science and Engineering, School of Engineering, The Hong Kong University of Science and Technology(计算机科学与工程系,香港科技大学);School of Computer Science and Technology, Huazhong University of Science and Technology(计算机科学与技术学院,华中科技大学)
高级计划与排程(Advanced Planning and Scheduling, APS)系统已成为现代制造运营不可或缺的部分,能在日益复杂动态的环境中实现资源优化配置与生产效率提升。尽管求解抽象调度问题的算法已得到广泛研究,但将制造需求转化为形式化约束这一关键前提仍依赖人工,且劳动密集度高。近年来生成式模型尤其是大语言模型(Large Language Models, LLMs)的进展,虽展现出从异构原始制造数据中自动生成约束规范的潜力,但直接应用仍面临自然语言歧义、输出非确定性以及领域特定知识有限等挑战。本文提出一种以约束为中心的架构,通过调控LLMs实现生产调度中可靠的约束自动规范。该架构定义了一个三级分层结构空间,借助领域特定表示实现,在保证精度与可靠性的同时维持灵活性。此外,本文设计并部署了一种自动化生产场景适配算法,可针对特定制造配置高效定制该架构。实验结果表明,所提方法成功平衡了LLMs的生成能力与制造系统的可靠性需求,在约束规范任务中显著优于纯LLM方法。
英文摘要:
Advanced Planning and Scheduling (APS) systems have become indispensable for modern manufacturing operations, enabling optimized resource allocation and production efficiency in increasingly complex and dynamic environments. While algorithms for solving abstracted scheduling problems have been extensively investigated, the critical prerequisite of specifying manufacturing requirements into formal constraints remains manual and labor-intensive. Although recent advances of generative models, particularly Large Language Models (LLMs), show promise in automating constraint specification from heterogeneous raw manufacturing data, their direct application faces challenges due to natural language ambiguity, non-deterministic outputs, and limited domain-specific knowledge. This paper presents a constraint-centric architecture that regulates LLMs to perform reliable automated constraint specification for production scheduling. The architecture defines a hierarchical structural space organized across three levels, implemented through domain-specific representation to ensure precision and reliability while maintaining flexibility. Furthermore, an automated production scenario adaptation algorithm is designed and deployed to efficiently customize the architecture for specific manufacturing configurations. Experimental results demonstrate that the proposed approach successfully balances the generative capabilities of LLMs with the reliability requirements of manufacturing systems, significantly outperforming pure LLM-based approaches in constraint specification tasks.
DyRo-MCTS: A Robust Monte Carlo Tree Search Approach to Dynamic Job Shop Scheduling
Centre for Data Science & Artificial Intelligence(数据科学与人工智能中心);School of Engineering & Computer Science(工程与计算机科学学院);Victoria University of Wellington(惠灵顿维多利亚大学)
Investigation of the Generalisation Ability of Genetic Programming-evolved Scheduling Rules in Dynamic Flexible Job Shop Scheduling
Centre for Data Science;Artificial Intelligence \& School of Engineering;Computer Science, Victoria University of Wellington, Wellington 6140, New Zealand
Deep Reinforcement Learning for Flexible Job Shop Scheduling with Random Job Arrivals
Centre for Artificial Intelligence, Zurich University of Applied Sciences(人工智能中心,苏黎世应用科学大学);Inspire AG(Inspire公司);Automatic Control Laboratory, ETH Zurich(自动控制实验室,苏黎世联邦理工学院)
Fairness-Aware Job Scheduling for Multi-Job Federated Learning
Nanyang Technological University(南洋理工大学);DSO National Laboratories(DSO国家实验室);Alibaba Group(阿里巴巴集团);Alibaba-NTU Singapore Joint Research Institute(阿里巴巴-新加坡南洋理工大学联合研究院)