IIoT-enabled digital twin for legacy and smart factory machines with LLM integration

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems Pub Date : 2025-04-03 DOI:10.1016/j.jmsy.2025.03.022

Anuj Gautam , Manish Raj Aryal , Sourabh Deshpande , Shailesh Padalkar , Mikhail Nikolaenko , Ming Tang , Sam Anand

{"title":"IIoT-enabled digital twin for legacy and smart factory machines with LLM integration","authors":"Anuj Gautam , Manish Raj Aryal , Sourabh Deshpande , Shailesh Padalkar , Mikhail Nikolaenko , Ming Tang , Sam Anand","doi":"10.1016/j.jmsy.2025.03.022","DOIUrl":null,"url":null,"abstract":"<div><div>Recent advancements in Large Language Models (LLMs) have significantly transformed the field of natural data interpretation, translation, and user training. However, a notable gap exists when LLMs are tasked to assist with real-time context-sensitive machine data. The paper presents a multi-agent LLM framework capable of accessing and interpreting real-time and historical data through an Industrial Internet of Things (IIoT) platform for evidence-based inferences. Real-time data is acquired from several legacy machine artifacts (such as seven-segment displays, toggle switches, and knobs), smart machines (such as 3D printers), and building data (such as sound sensors and temperature measurement devices) through MTConnect data streaming protocol. Further, a multi-agent LLM framework that consists of four specialized agents – a supervisor agent, a machine-expertise agent, a data visualization agent, and a fault-diagnostic agent is developed for context-specific manufacturing tasks. This LLM framework is then integrated into a digital twin to visualize the unstructured data in real time. The paper also explores how LLM-based digital twins can serve as real time virtual experts through an avatar, minimizing reliance on traditional manuals or supervisor-based expertise. To demonstrate the functionality and effectiveness of this framework, we present a case study consisting of legacy machine artifacts and modern machines. The results highlight the practical application of LLM to assist and infer real-time machine data in a digital twin environment.</div><div>Peer-review under responsibility of the scientific committee of the NAMRI/SME.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 511-523"},"PeriodicalIF":12.2000,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Manufacturing Systems","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0278612525000834","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, INDUSTRIAL","Score":null,"Total":0}

引用次数: 0

Abstract

Recent advancements in Large Language Models (LLMs) have significantly transformed the field of natural data interpretation, translation, and user training. However, a notable gap exists when LLMs are tasked to assist with real-time context-sensitive machine data. The paper presents a multi-agent LLM framework capable of accessing and interpreting real-time and historical data through an Industrial Internet of Things (IIoT) platform for evidence-based inferences. Real-time data is acquired from several legacy machine artifacts (such as seven-segment displays, toggle switches, and knobs), smart machines (such as 3D printers), and building data (such as sound sensors and temperature measurement devices) through MTConnect data streaming protocol. Further, a multi-agent LLM framework that consists of four specialized agents – a supervisor agent, a machine-expertise agent, a data visualization agent, and a fault-diagnostic agent is developed for context-specific manufacturing tasks. This LLM framework is then integrated into a digital twin to visualize the unstructured data in real time. The paper also explores how LLM-based digital twins can serve as real time virtual experts through an avatar, minimizing reliance on traditional manuals or supervisor-based expertise. To demonstrate the functionality and effectiveness of this framework, we present a case study consisting of legacy machine artifacts and modern machines. The results highlight the practical application of LLM to assist and infer real-time machine data in a digital twin environment.

Peer-review under responsibility of the scientific committee of the NAMRI/SME.

查看原文本刊更多论文

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Manufacturing Systems 工程技术-工程：工业

CiteScore

23.30

自引率

13.20%

发文量

216

审稿时长

25 days

期刊介绍： The Journal of Manufacturing Systems is dedicated to showcasing cutting-edge fundamental and applied research in manufacturing at the systems level. Encompassing products, equipment, people, information, control, and support functions, manufacturing systems play a pivotal role in the economical and competitive development, production, delivery, and total lifecycle of products, meeting market and societal needs. With a commitment to publishing archival scholarly literature, the journal strives to advance the state of the art in manufacturing systems and foster innovation in crafting efficient, robust, and sustainable manufacturing systems. The focus extends from equipment-level considerations to the broader scope of the extended enterprise. The Journal welcomes research addressing challenges across various scales, including nano, micro, and macro-scale manufacturing, and spanning diverse sectors such as aerospace, automotive, energy, and medical device manufacturing.