Digital twin designs with generative AI: crafting a comprehensive framework for manufacturing systems

被引:0
|
作者
Mata, Omar [1 ]
Ponce, Pedro [1 ]
Perez, Citlaly [2 ]
Ramirez, Miguel [2 ]
Anthony, Brian [3 ]
Russel, Bradley [3 ]
Apte, Pushkar [4 ]
Maccleery, Brian [5 ]
Molina, Arturo [1 ]
机构
[1] Tecnol Monterrey, Inst Adv Mat & Sustainable Mfg, Mexico City 14380, Mexico
[2] Tecnol Monterrey, Sch Engn & Sci, Mexico City 14380, Mexico
[3] MIT, Dept Mech Engn, Cambridge, MA 02139 USA
[4] Univ Calif Berkeley, AMPLab, 330 Sutardja Dai Hall,MC-1764, Berkeley, CA USA
[5] TechnoEvolutionAI, Austin, TX USA
关键词
Fuzzy TOPSIS; Morphological matrix; S4; framework; Digital twin; Large language model;
D O I
10.1007/s10845-025-02583-8
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The increasing prevalence of digital twins across various industries is attributed to their capacity for enhancing performance, minimizing downtime, and improving efficiency in the manufacturing process. These virtual models enable users to monitor and analyze the real-time behavior of physical assets, systems, or processes. However, the design of digital twins presents several challenges, such as excessive time costs, security risks, data accuracy, privacy concerns, and complexity in chosen components. This research paper introduces a comprehensive framework that effectively and flexibly addresses these challenges in digital twin development by employing S4 features, a Morphological Matrix, and Fuzzy TOPSIS. The proposed framework draws inspiration from methodologies utilized in product design and leverages a morphological matrix to identify and categorize diverse S4 features. These features are crucial in creating smart, sustainable, sensing, and socially impactful digital twins. Additionally, Fuzzy TOPSIS is employed to determine the optimal digital twin structure. The digital twin is divided into a physical and virtual model, interconnected and interactively communicated through a dedicated interface model. Furthermore, the framework integrates human knowledge into the digital twin structure, enhancing its robustness. When combined with a Large Natural Language Processing model, a comparative analysis of the digital twin structure is presented. A reconfigurable micromachine is used as a case study to validate the proposed approach. The results demonstrate the proposed framework's applicability for developing digital twins within the manufacturing industry. The main contribution of this paper is the innovative integration of human expertise and generative AI in a digital twin design framework. This approach provides a structured method that aligns the solution closely with industry requirements and offers a scalable and adaptable design methodology validated through practical applications. This framework is particularly valuable when dealing with a high number of devices that are challenging for human experts to evaluate rapidly in digital twin design.
引用
收藏
页数:24
相关论文
共 50 条
  • [21] Framework for extended digital manufacturing systems
    Nylund, Hasse
    Andersson, Paul H.
    INTERNATIONAL JOURNAL OF COMPUTER INTEGRATED MANUFACTURING, 2011, 24 (05) : 446 - 456
  • [22] Framework for Extended Digital Manufacturing Systems
    Nylund, Hasse
    Salminen, Kai
    Andersson, Paul
    PROCEEDINGS OF THE 6TH CIRP-SPONSORED INTERNATIONAL CONFERENCE ON DIGITAL ENTERPRISE TECHNOLOGY, 2010, 66 : 1483 - 1500
  • [23] A Comprehensive Review of AI-Based Digital Twin Applications in Manufacturing: Integration Across Operator, Product, and Process Dimensions
    Alfaro-Viquez, David
    Zamora-Hernandez, Mauricio
    Fernandez-Vega, Michael
    Garcia-Rodriguez, Jose
    Azorin-Lopez, Jorge
    ELECTRONICS, 2025, 14 (04):
  • [24] Proposing a Small-Scale Digital Twin Implementation Framework for Manufacturing from a Systems Perspective
    Loaiza, Jonatan H.
    Cloutier, Robert J.
    Lippert, Kari
    SYSTEMS, 2023, 11 (01):
  • [25] A Unified Digital Twin Framework for Real-time Monitoring and Evaluation of Smart Manufacturing Systems
    Qamsane, Yassine
    Chen, Chien-Ying
    Balta, Efe C.
    Kao, Bin-Chou
    Mohan, Sibin
    Moyne, James
    Tilbury, Dawn
    Barton, Kira
    2019 IEEE 15TH INTERNATIONAL CONFERENCE ON AUTOMATION SCIENCE AND ENGINEERING (CASE), 2019, : 1394 - 1401
  • [26] A DIGITAL TWIN ASSISTED FRAMEWORK FOR QUALITY ASSURANCE IN MOULD MANUFACTURING
    Boettjer, Till
    Ramanujan, Devarajan
    PROCEEDINGS OF ASME 2023 INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE, IDETC-CIE2023, VOL 2, 2023,
  • [27] Generative Design and Digital Manufacturing: Using AI and robots to build lightweight instruments
    McClelland, Ryan
    CURRENT DEVELOPMENTS IN LENS DESIGN AND OPTICAL ENGINEERING XXIII, 2022, 12217
  • [28] Design and operations framework for the Twin Transition of manufacturing systems
    van Erp, T.
    Rytter, N. G. M.
    ADVANCES IN PRODUCTION ENGINEERING & MANAGEMENT, 2023, 18 (01): : 92 - 103
  • [29] Digital twin and IOT technology for secure manufacturing systems
    Gopal L.
    Singh H.
    Mounica P.
    Mohankumar N.
    Challa N.P.
    Jayaraman P.
    Measurement: Sensors, 2023, 25
  • [30] Optimizing smart manufacturing systems using digital twin
    Ojstersek, R.
    Javernik, A.
    Buchmeister, B.
    ADVANCES IN PRODUCTION ENGINEERING & MANAGEMENT, 2023, 18 (04): : 475 - 485