4,382
Views
10
CrossRef citations to date
0
Altmetric
Editorial

Digital twin-enabled smart industrial systems: recent developments and future perspectives

ORCID Icon, ORCID Icon, &

1. Introduction

A digital twin refers to a virtual representation of a physical asset or system throughout its lifecycle (Mikell and Clark Citation2018). It leverages the Internet of Things (IoT) to facilitate learning, reasoning, and improving decision-making in complex systems (Rosen et al. Citation2015). The adoption of digital twin technologies is becoming more prevalent in the manufacturing industry, including product design, maintenance, and production planning (Tao et al. Citation2018a, Citation2018B, Citation2018C). Its concept has also led to a variety of emerging applications in other industries, such as energy (e.g. Iglesias et al. Citation2017), transportation (e.g. Gregory, Leonard, and Scotti Citation2016), and healthcare (e.g. Bruynseels, Santoni de Sio, and van den Hoven Citation2018). The core of the technologies consists of artificial intelligence algorithms, enabled by machine learning methods trained on massive amounts of data captured through numerous connected sensors on the physical objects (Kuo and Kusiak Citation2019). Computer simulations are run to examine different possible scenarios to predict the outcomes of decisions. For timely interaction, digital twins evolve and continuously update to reflect changes in their physical counterparts (Siemens Citation2021). The convergence between the physical assets and the virtual greatly determines the success of the applications (Tao and Zhang Citation2017; Bortolini et al. Citation2018). The ultimate goal of digital twins is to drive systems smarter and be more efficient.

While digital twin technologies have shown promising benefits, there are still challenges in their implementations, for example, synchronization between physical assets and virtual space, optimization and coordination on connected networks, dealing with uncertainty in systems, and development of computationally efficient algorithms for real-time response. New mathematical modeling techniques and solution methodologies, powered by new types of auto-sensed data, are needed to tackle the challenges. The emergence of digital twin technologies has presented new research opportunities.

2. About this special issue

The purpose of this Special Issue of the International Journal of Computer Integrated Manufacturing is to present the state-of-the-art research on smart industrial systems enabled by digital twin technologies, to demonstrate the benefits of the adoption of the technologies in complex systems and to anticipate the potential challenges.

The Guest Editors considered submissions that introduce new research problems and concepts, develop novel and rigorous methodologies to tackle the problems, and present innovative applications. The submissions received present original and high-quality research on digital twin-enabled smart industrial systems. After a regular and rigorous review process, 13 submissions were accepted that contribute to the research and development of digital twin applications in the following aspects:

Bibliometric review

The first paper in this special issue aims to outline the state-of-the-art of digital-twin applications in smart industrial systems. The paper, Digital twin-enabled smart industrial systems: a bibliometric review, by Ciano et al. investigates the body of literature on digital twins and explores, in particular, their role in enabling smart industrial systems. This review adopts a dynamic and quantitative bibliometric method, including works citations, keywords co-occurrence networks, and keywords burst detection with the aim of clarifying the main contributions to this research area and highlighting prevalent topics and trends over time.

Smart manufacturing

Smart manufacturing is a key area that digital twin technologies have been applied. Information about the status of different manufacturing processes could be captured in real time. Decisions can be determined in the virtual world and delivered for automation in the physical manufacturing system. Zhang et al. present a five-dimensional fusion model with the use of robotics to support automatic reconfiguration in their paper, titled Digital twin-enabled reconfigurable modeling for smart manufacturing systems. Their proposed model can represent the manufacturing resources in the physical world and also incorporate the capabilities and dependencies of the digital twins. The authors develop a prototype that shows the operational efficiency gain due to the proposed system.

The third paper, Wafer sojourn time fluctuation analysis for time-constrained dual-arm multi-cluster tools with activity time variation, by Yang et al. considers the scheduling of multi-cluster tools in semiconductor manufacturing systems. The problem is challenging as activity times vary. They propose a two-level real-time operational architecture and a real-time control policy to compute the limit of wafer sojourn time delay in a process chamber.

Zheng et al. present an online inspection system in their paper, Digital twin for geometric feature online inspection of car body-in-white. They propose a three-level virtual modeling approach with an Element-Behavior-Rule for modeling the physical space. With an emulator experimental environment, they demonstrate that the inspection process can be monitored in real time, and the proposed approach is feasible for practical use.

The fifth paper, A digital twin-enabled value stream mapping approach for production process reengineering in SMEs, by Lu et al. introduces integration of the Internet of Things (IoT) technology and Efficiency Validate Analysis (EVA) simulation framework. This paper proposes a digital twin-enabled Value Stream Mapping (VSM) approach for small and medium enterprises (SMEs). This easy-to-adopt and accurate method helps manufacturing SMEs in redesigning and reengineering production processes. This study contributes to research on digital twin modeling, by proposing a theoretical framework that combines IoT data-driven production process planning and simulation methods.

Smart logistics

After products are manufactured, they will be delivered to the customers or consumers through a logistics system. Digital twin technologies have driven logistics systems to be more efficient and responsive. Leng et al. present in their paper, Digital twin-driven joint optimisation of packing and storage assignment in large-scale automated high-rise warehouse product-service system, a digital twin system that utilizes real-time data from a physical warehouse product-service system to construct a cyber model. The information is used to enable an integrated optimization of stacked packing and storage assignments. They develop a conduct a case study of a tobacco warehouse and demonstrate that the proposed approach could maximize the utilization and efficiency of a large-scale automated high-rise warehouse.

Wong et al. in the seventh paper propose a closed-loop digital twin system for air cargo load planning operations. Their system integrates cargo loading plan optimization and simulation models, virtual reality (VR), IoT, real-time sensing technologies to connect, monitor, and control operations in both physical and virtual spaces related to air cargo operations. The authors develop a Cave Automatic Virtual Environment (CAVE)-based VR system to visualize the cargo loading operations and conduct experiments. A feedback loop with data captured by sensors helps to suggest decisions on optimal air cargo loading plans.

Process synchronization is an essential aspect of an efficient logistic system. Pan et al. propose a digital-twin-driven production logistics synchronization system for vehicle routing problems with pick-up and delivery in industrial park. The authors build a digital-twin architecture to support decision-making for vehicle routing in an industrial park. They present a real-time dynamic synchronization control mechanism to synchronize transportation operations in the production logistics system.

Smart energy management

Energy management is an important component in all industrial systems. Applications of digital twin technologies have been developed for more efficient use of energy. Ma et al. propose a case-practice-theory-based (CPTB) method in their paper, Acase-practice-theory-based method of implementing energy management in amanufacturing factory, to assist energy management in amanufacturing factory. Their goal is to reduce energy and emissions. They utilize IoT, cyber-physical systems, and big data to enable their application. Their results show that the proposed method could reduce the energy consumption of the production by 3% and the energy costs by 4%.

In the tenth paper, Barenji et al. propose a digital twin approach for robotic cellular in their paper, titled A digital twin-driven approach towards smart manufacturing: reduced energy consumption for a robotic cell. In their work, a real-time motion planning decision-making agent is developed to minimize energy consumption. They establish a linkage between the physical and virtual objects in a manufacturing workshop to enable autonomous decision -making and conduct a simulation analysis. Their study demonstrates the good performance of their proposed approach in terms of energy consumption.

Smart human-machine interaction

One important aspect of digital twin applications is their ability to interact with humans and learn from human actions for more effective applications. The first step is to understand human behavior from data. In the paper Digital twins in human understanding: a deep learning-based method to recognize personality traits, Sun et al. use digital twin models to capture human personalities by studying the content users posted and liked on social media. They propose a multi-task learning deep neural network approach to inferring human personality with social media posts and likes. Their experimental results suggest that these two types of information are useful in enhancing the performance of their approach.

The paper, User acceptance of virtual reality technology for practicing digital twin-based crisis management, by Kwok et al. proposes a research model to examine the effects of perceived usefulness, perceived ease of use, perceived behavioural control, application-specific self-efficacy, and attitude on users acceptance of the virtual reality systems in the context of training for crisis management in the manufacturing environment.

Wang et al. in the thirteenth and final paper, present an application of digital twin for human-machine interaction with convolutional neural network. They apply deep learning to enable human-machine interaction by utilizing both physical and virtual data. They propose modified 3D-Visual Geometry Group Network and 3D-Residual Network models to learn from humans’ actions captured in videos. Their experimental results demonstrate the good performance of the proposed approach on human motion recognition.

3. Challenges and future opportunities

The contributions of this special issue not only present the state-of-the-art of digital twin technologies but also suggest several challenges and future research opportunities.

Synchronization of numerous objects in the physical world and entities in the virtual space

Synchronization is a core element of digital twin technologies. The first type of synchronization is between the physical objects and entities in the virtual space. This type of synchronization requires effective and reliable sensing technologies to capture information about the physical objects and fast communications between sensors and the digital twin platform. The second type of synchronization is between the entities in the virtual space. This requires consistent data collection frequency for building or updating the entities in the virtual space, such that the interactions between the entities in the virtual world would reflect the physical phenomenon in reality. Ultimately, digital twin technologies can help enable the third type of synchronization regarding the operations in the real world, as demonstrated in Pan et al. in this issue.

Digital twin for full product life cycle

The contributions of this special issue focus on some specific applications of digital twin in a product life cycle (e.g. manufacturing, energy control, and logistics). As Zhang et al. in their paper suggest, digital twin-enabled life cycle models, which utilize big data collected at different stages of the product life cycle, would be essential to address the exact needs for product development and service design.

User acceptance of digital twin technologies

The majority of studies aim to address the technical challenges of the development of digital twin technologies. There is relatively little research on the user acceptance of the technologies. The paper by Kwok et al. in this issue, demonstrates how digital twin technologies can be improved from the user’s point of view. The data collected from the virtual reality environment can be used to understand the user’s attitude towards the technology and be leveraged to enhance the design of the digital twin platform. As user experience is a critical element to determine if the technology will be deployed, research on user acceptance of digital twin is essential to support its actual implementation in the industry.

Utilization of social media data for the development of digital twin

Many of the applications of digital twin technologies emphasize the representation of physical objects in a virtual world. The paper by Sun et al. in this issue, shows that understanding human behavior is also essential for more effective decision-making in a digital twin application. Their study demonstrates that social media could be a good source of information to incorporate human personality and preference in the development of digital twin. The utilization of users’ personality traits may also facilitate product design and production planning.

Digital twin technologies for small and medium-sized enterprises

While the contributions in this special issue have demonstrated the effectiveness and promising applications of digital twin technologies, as suggested in the paper by Barenji et al. in this issue, the deployment of the technologies in small and medium-sized enterprises (SMEs) is economically challenging. Upgrading and replacing the existing physical workshops and production systems would require a high investment, and the return of such an investment is not guaranteed. The impacts of the new technologies on the businesses and operations of SMEs are also not clear before the deployment. Collaboration between large corporates and SMEs may be a solution to this problem.

Integrated digital twin of separate existing systems

Most of the existing studies developed brand-new digital twin systems from scratch for their applications. However, for large-scale corporates or organizations with a long history, it may not be possible to build a new digital twin platform to replace their existing systems. Development of an integrated digital twin platform that connects separate existing systems is challenging (Shen, Wang, and Deng Citation2021). Heterogeneous data format, granularity, quality, and collection frequency across various information systems result in data inconsistencies. A unified framework for data transformation and imputation would be a solution to establish such an integrated digital twin platform.

Establishment of digital twin cities

In many cities around the globe, smart city initiatives have been outlined to optimize the usage of resources, boost the competitiveness of the city, and ultimately enhance citizens’ quality of life. As demonstrated by the contributions of this special issue, applications of digital twin technologies span multiple industries, including manufacturing, logistics, and energy. Deng, Zhang, and Shen (Citation2021) propose the concept of a digital twin city, where all entities in such a digital city would have historical records that can be traced, the present state of each entity can be checked, and prediction of future states can be made. Digital twin technologies deployed in various industries would be the backbone to support and advance the establishment of digital twin cities.

4. Conclusions

With the advance of sensing and computing technologies, applications of digital twin have been more promising and prevalent. This special issue of International Journal of Computer Integrated Manufacturing presents the state-of-the-art digital-twin technologies and their applications in various aspects, including smart manufacturing, smart logistics, smart energy management, and smart human-machine interaction. We hope that this special issue can stimulate future research on the development of digital twin technologies and lead to further innovative and impactful applications.

Acknowledgments

The Guest Editors would like to express their sincere gratitude to Professor Stephen Newman, Editor-in-Chief, and Professor Aydin Nassehi, Senior Editor, of International Journal of Computer Integrated Manufacturing, for the opportunity to organize this special issue and their advice and suggestions. The Guest Editors also thank all the contributors to this special issue and the reviewers for their time and effort. Further acknowledgement is to the partial support from the 2019 Guangdong Special Support Talent Program – Innovation and Entrepreneurship Leading Team (China) (2019BT02S593) for the work related to the organization of this special issue.

Disclosure statement

No potential conflict of interest was reported by the author(s).

References

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.