TY - JOUR
T1 - Nominal digital twin for new-generation product design
AU - Zhang, Haizhu
AU - Li, Rong
AU - Ding, Guofu
AU - Qin, Shengfeng
AU - Zheng, Qing
AU - He, Xu
N1 - Funding information: This work was supported by the National Key Research and Development Program of China (grant number 2020YFB1708000), Sichuan Science and Technology Support Project (grant numbers 2021YFG0039 and 2022YFG0252), and National Natural Science Foundation of China (grant number 52105277).
PY - 2023/9/1
Y1 - 2023/9/1
N2 - With the increasing competition of the manufacturing industry, it is essential for manufacturers to develop a new (or next)-generation product based on their prior product design, product user experience, historical performances, etc. The digital twin (DT) is believed to be a suitable technology to support product lifecycle management with excellent data capture and analysis capabilities and product family design capabilities. However, it remains a challenge to synthesize and incorporate the data and information captured from previous generational product development and stored on their individual digital twin instances (DTIs) into next-generation product design. To address this problem, this study proposes a new nominal digital twin (NDT) concept as a collective digital representation of the current generational product digital mockups (DMUs) and all individual DTIs of the built products in services for new-generational product development. NDT is first defined here as a prototypical and synthesized digital twin (or digital representation) of multiple individual digital twins corresponding to multiple physical products in use or previously used. By analyzing the data and information on their DTIs, NDT can enable the establishment and evolution of a more precise approximated model of many related family products used previously or currently in use in various application scenarios and environments in the physical world. This paper also demonstrates how a NDT model can be first established in the product design phase from various digital mockup models and enhanced later with a stochastic forest meta-model based on Bayesian optimization connected to DTIs. With this NDT model, collaborative exploration for optimal design solutions during new-generation product design and improvement can be performed on NDT through multi-objective optimization, which in turn can make new-generation product design easier and quicker. As a primary verification of the feasibility of our proposed approach, a case study has been carried out, and the results have well confirmed that the NDT-based new-generation product design approach is feasible.
AB - With the increasing competition of the manufacturing industry, it is essential for manufacturers to develop a new (or next)-generation product based on their prior product design, product user experience, historical performances, etc. The digital twin (DT) is believed to be a suitable technology to support product lifecycle management with excellent data capture and analysis capabilities and product family design capabilities. However, it remains a challenge to synthesize and incorporate the data and information captured from previous generational product development and stored on their individual digital twin instances (DTIs) into next-generation product design. To address this problem, this study proposes a new nominal digital twin (NDT) concept as a collective digital representation of the current generational product digital mockups (DMUs) and all individual DTIs of the built products in services for new-generational product development. NDT is first defined here as a prototypical and synthesized digital twin (or digital representation) of multiple individual digital twins corresponding to multiple physical products in use or previously used. By analyzing the data and information on their DTIs, NDT can enable the establishment and evolution of a more precise approximated model of many related family products used previously or currently in use in various application scenarios and environments in the physical world. This paper also demonstrates how a NDT model can be first established in the product design phase from various digital mockup models and enhanced later with a stochastic forest meta-model based on Bayesian optimization connected to DTIs. With this NDT model, collaborative exploration for optimal design solutions during new-generation product design and improvement can be performed on NDT through multi-objective optimization, which in turn can make new-generation product design easier and quicker. As a primary verification of the feasibility of our proposed approach, a case study has been carried out, and the results have well confirmed that the NDT-based new-generation product design approach is feasible.
KW - Closed-loop iterative
KW - Digital mock-up
KW - Digital twin-based optimal design
KW - New-generation product design
KW - Nominal digital twin
UR - http://www.scopus.com/inward/record.url?scp=85165696768&partnerID=8YFLogxK
U2 - 10.1007/s00170-023-11924-x
DO - 10.1007/s00170-023-11924-x
M3 - Article
AN - SCOPUS:85165696768
VL - 128
SP - 1317
EP - 1335
JO - International Journal of Advanced Manufacturing Technology
JF - International Journal of Advanced Manufacturing Technology
SN - 0268-3768
IS - 3-4
ER -