Large language models in intelligent manufacturing and mechanical engineering: a review of robotics, fault diagnosis, design, and engineering knowledge workflows
Jul 2026· Journal of Intelligent Manufacturing· 0 citations· 67 references
TL;DR
The evidence indicates that LLMs are becoming useful semantic and coordination layers in engineering workflows, but not dependable engineering substitutes in human-in-the-loop, evidence-grounded systems where retrieval, validation, tool use, and structured knowledge help keep outputs useful and bounded in safety-relevant tasks.
Abstract
Large language models (LLMs) are attracting growing attention in robotics, mechanical engineering, and mechatronic systems. This review shows that, in most engineering settings, their value is not that they replace simulation, control, or numerical analysis tools, but that they help engineers work across documents, data sources, software environments, and natural-language instructions more efficiently. The review surveys recent studies across robotics and embodied systems, fault diagnosis and maintenance, design and simulation, industrial knowledge workflows, manufacturing knowledge systems, and selected adjacent engineering applications only where they provide transferable methodological insight for intelligent manufacturing. Across these areas, the evidence shows a movement away from prompt-only demonstrations and toward retrieval-augmented generation (RAG), multimodal, tool-connected, and agent-based systems that are more tightly grounded in engineering evidence and operational context. The review further shows that the strongest results generally come from hybrid architectures that combine LLMs with RAG, knowledge graphs, multimodal perception, digital twins, validation modules, or downstream engineering tools. Even so, important limitations remain, including weak grounding in cluttered or ambiguous settings, limited numerical and spatial reliability, poor long-horizon robustness, fragmented benchmarks, and incomplete integration with trusted engineering software. Overall, the evidence indicates that LLMs are becoming useful semantic and coordination layers in engineering workflows, but not dependable engineering substitutes. Their most credible near-term role is in human-in-the-loop, evidence-grounded systems where retrieval, validation, tool use, and structured knowledge help keep outputs useful and bounded in safety-relevant tasks.
This review examines recent progress in large AI models for intelligent manufacturing, covering model architectures, adaptation strategies, system integration, and applications across product development, production processes, equipment maintenance, and manufacturing services.
Baotong Chen, Lu Dai, Chuangjian Wang et al.· IEEE Access· 0 citations
Generative artificial intelligence (GAI) is expanding from model-centered research into engineering and manufacturing activities, but its scope and maturity remain uneven. This PRISMA-guided bibliometric and abstract-level thematic review maps peer-reviewed industrial GAI research published from 2022 to 4 June 2026. Searches of Scopus, Web of Science, and the ACM Digital Library identified 492 records; 119 duplicates and 121 ineligible records were removed, leaving 252 studies. Keyword normalization, co-occurrence analysis, dominant and secondary thematic coding, and an abstract-reported evidence characterization were applied. The corpus shows two connected trajectories: engineering generation based on generative models for design, topology, materials, and electronics, and knowledge-intensive industrial intelligence based on large language models, retrieval-augmented generation, knowledge graphs, agents, and human–AI collaboration. Most studies report empirical or computational evaluation (72.2%), but 84.5% remain research-stage; only 0.8% indicate operational industrial evidence in their abstracts. The findings, therefore, distinguish publication activity from deployment maturity. Priority requirements for adoption include domain-grounded data, verification, manufacturability checks, traceability, cybersecurity, intellectual property protection, system integration, workforce preparation, and human accountability. This review contributes a reproducible cross-domain map, an overlap-aware synthesis, and stakeholder-specific guidance for trustworthy industrial GAI.
Robots are being deployed for an increasingly diverse set of purposes, from industrial manufacturing to delivery, inspection, and surgical assistance, and the systems entering these roles are markedly more capable than earlier generations, utilizing learned perception, language-based planning, and multi-sensor input. This increase in the number of deployments is reflected in industry forecasts that report rapid, sustained growth in industrial robot installations and in the worldwide operational stock [18]. As robots take on broader and more complex missions in human-centered environments, their quality assurance and physical safety become increasingly relevant. However, despite this growth and advancements, few existing works offer a comprehensive review of robotic test automation with its conventional, AI-powered, and agentic landscape and overall trends. This paper addresses this gap by summarizing, classifying, and visualizing conventional and AI-based robot testing. Extending this, a reading of the commercial landscape suggests that conventional and AI methods act as complements rather than substitutes. Lastly, this work portrays many problems, challenges, and needs to aid in future research.
Karthik Rengarajan, Siddharth Pokuri, J. Gao· International Conference on...· 0 citations
Building disassembly is critical for circular economy material reuse, yet remains rare due to cost and safety constraints, leading to demolition and material downcycling. Automation could improve both efficiency and safety, but currently available technology does not yet enable full automation. We propose a human–robot collaboration system architecture that uses agentic large language models. We test this approach in building disassembly—an unstructured, safety–critical domain where conventional pre-programmed robotics are inadequate. The agentic architecture combines curated domain knowledge, physics simulation for stability validation, and natural language interfaces, enabling the robot to participate through proactive reasoning rather than follow control commands. We evaluated the architecture through three progressively complex scenarios: collaborative spatial adaptation, collaborative decision-making, and learning. The main contribution is a modular, data-grounded HRC methodology in which specialized LLM agents perform agentic reasoning: the robot assesses situations, retrieves relevant procedural knowledge, validates decisions through simulation, and negotiates solutions with human operators. This proof of concept demonstrates that agentic multi-agent LLM systems can enable adaptive human–robot collaboration under uncertainty, beyond natural language interfaces through integrated domain knowledge, physics validation, and agentic reasoning.
Samuel Slezák, Shirin Shevidi, Zahra Shakeri et al.· Construction Robotics· 0 citations
Industrial operations need AI systems that can reason across live process data, engineering knowledge, and operator workflows. Yet conventional machine learning models often remain narrow predictors, while large language models lack grounding in plant behaviour, constraints, and real-time operating context. This talk presents Orbital, a grounded multi-agent system for decision support in industrial operations. Orbital combines three complementary layers: a time-series model for multivariable process dynamics and uncertainty-aware forecasting; a constraint-learning layer that extracts engineering relationships from plant documentation, including P&IDs, datasheets, mass and energy balances, and operating manuals; and a language-fusion layer that aligns process behaviour with engineering descriptions. These components are coordinated through specialist agents for planning, tool execution, verification, memory, and response composition. The system moves beyond prediction toward interpretable decision support: detecting abnormal behaviour, retrieving relevant historical events, explaining likely root causes, and grounding recommendations in both data and engineering constraints. More broadly, this work argues that the next generation of industrial AI must be grounded, multi-modal, and operationally trustworthy; connecting data, domain knowledge, and human decision-making in high-consequence environments.
Samyakh Tukra· Proceedings of the 3rd Found...· 0 citations
Through applied case studies in pharmaceutical discovery and financial systems, common design patterns that make agentic systems successful are analyzed, and practical mitigation strategies for failure modes are discussed, such as verification pipelines, fallback mechanisms, and human-in-the-loop supervision.
Grace Hui Yang, P. Venkit, Hooman Sedghamiz et al.· Proceedings of the 32nd ACM...· 0 citations