发表机构
Singapore Institute of Manufacturing Technology (SIMTech), Agency for Science, Technology and Research (A*STAR); Advanced Remanufacturing and Technology Centre (ARTC), Agency for Science, Technology and Research (A*STAR); School of Mechanical and Aerospace Engineering, Nanyang Technological University（新加坡制造技术研究院（新加坡科学、技术与研究局）; 先进再制造与技术中心（新加坡科学、技术与研究局）; 南洋理工大学机械与宇航工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出Design-to-Plan多智能体框架，结合LLM与确定性模块，实现从3D CAD及2D图纸到制造工艺规划的端到端自动化，经300个基准案例验证，性能优异且令牌用量显著降低。
AI 中文摘要
制造工艺规划将异构设计信息转化为连贯的制造决策。然而，现有方法聚焦于孤立子任务，如特征识别、图纸解读或刀具选择，难以支持从设计工件到工艺规划的完整推理链，而当规划需解读3D CAD模型、2D工程图、材料及领域特定规则时，该推理链至关重要。为解决此缺口，本文提出Design-to-Plan，一种基于大语言模型（LLM）的多智能体端到端制造工艺规划框架。该框架由一个协调器，协调负责3D特征识别、2D图纸分析、2D-3D上下文融合、知识检索、工艺排序、刀具选择及报告生成的专用智能体。与将LLM作为独立文本生成器的做法不同，本框架将其部署为推理智能体，与确定性模块及知识源交互以生成一致且可追溯的决策。在这种混合设计中，确定性模块和专用智能体从CAD及图纸输入中提取结构化信息，而LLM智能体执行上下文感知推理、检索制造规则、解决冲突并生成规划输出。本框架通过300个基准案例，对三个下游启用ReAct的智能体进行评估，同时单独评估CAD特征识别、图纸分析及2D-3D上下文融合。该并行架构在下游智能体中实现100%成功率，刀具F1分数为95.9%-97.6%，冲突分析中源检测准确率为90%，关键规划任务的令牌使用量减少60%-68%。结果表明，基于LLM的结构化多智能体协调可弥合设计表示与制造知识，实现可扩展、高效且可追溯的设计到规划自动化。
英文摘要
Manufacturing process planning transforms heterogeneous design information into coherent manufacturing decisions. However, existing approaches focus on isolated subtasks, such as feature recognition, drawing interpretation, or tool selection, and struggle to support the full reasoning chain from design artifacts to process plans. This is critical when planning must interpret 3D CAD models, 2D engineering drawings, materials, and domain-specific rules. To address this gap, this paper presents Design-to-Plan, a large language model (LLM)-based multi-agent framework for end-to-end manufacturing process planning. An orchestrator coordinates specialized agents for 3D feature recognition, 2D drawing analysis, 2D-3D context fusion, knowledge retrieval, process sequencing, tool selection, and report generation. Rather than using LLMs as standalone text generators, the framework deploys them as reasoning agents that interact with deterministic modules and knowledge sources to produce consistent and traceable decisions. In this hybrid design, deterministic modules and specialized agents extract structured information from CAD and drawing inputs, while LLM agents perform context-aware reasoning, retrieve manufacturing rules, resolve conflicts, and generate planning outputs. The framework is evaluated using 300 benchmark cases across three downstream ReAct-enabled agents, plus separate evaluations of CAD feature recognition, drawing analysis, and 2D-3D context fusion. The parallel architecture achieves 100% success across downstream agents, Tool F1 scores of 95.9%-97.6%, 90% source detection accuracy in conflict analysis, and a 60%-68% reduction in token usage for key planning tasks. Results show that structured LLM-based multi-agent coordination can bridge design representations and manufacturing knowledge, enabling scalable, efficient, and traceable design-to-plan automation.
CommentsSubmitted to Elsevier Journal
