Digital Twin: A Quick Overview

In this overview, we will discuss the definition of “digital twin” and the relationship between “digital twin” and related terms often used in the same discussion. Many definitions of “digital twin” proposed throughout the engineering community are listed in Table 1 of the Appendix. The commonality they share are the following features, conveyed either implicitly or explicitly:

• A digital twin is a digital model
• A digital twin represents a specific real-world physical asset
• A digital twin is kept up to date with changes to the real-world physical asset.

Combined, the digital twin framework includes a digital model (twin), its corresponding physical asset, and a data connection between the two as shown in Figure 1.

Figure 1: Physical and Digital Environment

While not always agreeing on the details, the community suggests that digital twins have three variable characteristics:

• Fidelity of the digital model
• Direction of information flow between the digital twin and corresponding physical asset
• Timeliness of the data flowing between the digital twin and corresponding physical asset.

Thus, the most basic definition of a digital twin must reflect that it is a digital representation of a real-world physical asset with parameters that update with sufficient rapidity to enable practical applications as the physical asset changes.  The definition of model is a representation of a system of interest for a specific purpose.  It follows that the most basic definition of a digital twin is:

A digital representation of a specific real-world system of interest for a specific purpose

This definition contains all the common features listed above but also allows for additional features if the specific purpose requires them to add value. The clause “for a specific purpose” guides us by preventing “modeling for modeling’s sake” or in this case, “digital twinning for digital twinning’s sake.”

We extend the most basic definition to include the variable characteristics, resulting in ITEA’s definition of a digital twin for the T&E community:

We also point out that the boundary that defines the “system” changes based on the use case. A digital twin may include what some consider “environment” in other use cases. For example, in T&E, a digital twin may include a digital representation of some or all of a test range. The flexibility of a use case sensitive definition of “system” allows us to focus on what digital twins are intended to do: provide value by allowing some use cases to take advantage of the infinitely composable digital world rather than the more constrained physical world.

Digital twins are distinguished from digital threads. The Defense Acquisition University (DAU) defines a digital thread as:

An extensible, configurable, and component, enterprise-level analytical framework that seamlessly expedites the controlled interplay of authoritative technical data, software, information, and knowledge in the enterprise data-information-knowledge systems, based on the Digital System Model template, to inform decision makers throughout a system’s life cycle by providing the capability to access, integrate, and transform disparate data into actionable information.
Digital threads can and should be used to develop models and digital twins. The diagram below shows the relationship among a digital twin, its corresponding physical asset, and a digital thread.

Figure 2: Digital Twin, Corresponding Physical Asset, and Digital Thread

The digital thread provides stakeholders with seamless access to tools and an authoritative source of truth for the technical information pertaining to the physical asset and its operational environment.

A digital model is typically a manually updated system representation implemented at the type or class level whereas a digital twin is an automatically updated representation that is implemented at the specific unit level to dynamically respond to the changes in the physical world. Figure 3 below illustrates these manual and automatic data flows between physical and digital objects.

Figure 3: Physical and Digital Manual and Automatic Data Flows

Digital shadows are another type of model, differing from digital twins only in that they have manual data flows to their physical counterpart.
A digital twin then is the most advanced model and, like a digital shadow, is a tailored digital representation of not just the class type of object it represents but rather the specific individual unit of that type (e.g., specific hull or tail number). The distinguishing factor between digital twins and digital shadows is that automated data flows are bi-directional in the former but not the latter. This allows for changes and updates to digital twins to be tested in digital environments and then pushed in real-time to their physical counterparts. Automatic data flows allow for more informed decisions, updates, and improvements of both the digital and physical objects, as well as autonomous scenario execution.

This section describes how digital models and twins can provide value across the entire acquisition program lifecycle of a system, including the phases of design and development, integration, T&E, and operations and sustainment.

Figure 4 illustrates the Digital Twin as a virtual representation of a connected physical asset and is borrowed from “Digital Twin: Definition & Value,” an Aerospace Industries Association (AIA) position paper.

Figure 4. AIAA Representation of a Digital Twin

During system design and development, basic technological components are integrated to establish that they will work together. The conceptual system model is translated into schematics and diagrams, including model-based systems engineering simulations that are consistent with the intended use of the model. Hardware and software specifics are defined. Information about how the model and simulations are organized, constructed, and executed are generated. The model development plan is published, outlining basic assumptions, capabilities, and limitations and risks associated with system development, testing, and operations and sustainment. Digital twins of early physical prototypes can be implemented and increased in fidelity as system development progresses; these twins can be exposed to operational contexts in virtual environments much earlier than physical prototypes can be in the physical environment, enabling earlier identification of deficiencies as well as their corrections. The required level of digital twin maturity and fidelity needed to support the system across its lifecycle is defined; in many cases only high-level system data is needed to address questions of interest and so it is important to remember that digital twins are not one-size-fits-all solutions. Indeed, digital twins should be built for specific applications and users.

During the integration phase, models are used to assess system, sub-system, and component-level elements for interface compatibility and data exchange interoperability. Models during this phase are valuable for physical and logical integration and test planning. Ultimately, the goal of integration is to ensure that the individual system elements function properly as a whole and that the system’s technical requirements and operational capabilities are satisfied. The diagram below shows the bottom-up branch of the Vee Model, including the tasks of assembly, integration, element-level verification during integration, and system-level validation; data flows between the physical and digital domains throughout this process.

Figure 5: Integrated Models Interactions During Integration Phase

Comprehensive live testing of future warfighting capabilities will not be possible in many cases due to environmental, fiscal, safety, classification, or ethical constraints, and so T&E will become more dependent on digital representations to test the performance and interoperability of our systems.
By taking advantage of the composability, editability, and parallelism of digital representations, digital twins can be used to augment T&E by replacing physical testing with virtual testing. In the future, we envision using digital twins alongside other digital models to not only provide a thorough understanding of the operational performance of single systems, but also to glean the high-level and synergistic mission-level effects characteristic of future joint warfighting operations, as well as the emergent behaviors they entail inside of “digital arenas”. This same method can be extended to support tactical decision-making during real-world combat operations in real-time.

To achieve this, however, digital twins must pass verification, validation, and accreditation (VV&A) prior to their use in T&E, following all relevant DoD policy on the use of M&S and its VV&A. In doing so, quantitative estimates of model error or uncertainty are determined with an emphasis on enabling rigorous interpolation and extrapolation beyond the parameter space or operational envelope that the digital twins have been directly validated on.
Program managers should plan early for the investment required to develop and validate digital twins while carefully considering the expected limitations of live testing that digital twins will be able to mitigate.

As described in the previous section, digital twins and other models can be injected into virtual environments to create mirrors of real-world combat operations, enabling intelligent data-driven tactical decision-making in real-time.

During sustainment, decisions regarding the models’ maintenance, disposal, disposition, etc. are made. These decisions include how much a system costs to maintain and how much its capability depreciates over time relative to the current technological landscape. One valuable application of digital twins during this phase is using them to predict when system maintenance tasks need to be performed; data from the physical system can be analyzed alongside the system’s maintenance log to predict when maintenance tasks are required, helping to prevent downtime and reduce costs.

Digital twins are revolutionizing how industries approach product design and development, T&E, and operations and sustainment by bridging the gap between the physical and digital worlds. They are enabling organizations to make more informed decisions, reduce risks, and maximize the efficiency of their processes and products across their lifecycles while achieving deeper understanding of systems’ performance. As programs mature across their lifecycles, so too do their digital twins—beginning as primarily simple, static digital models that evolve over time into twins that can exchange operational data with their physical counterparts in real-time.

This overview has provided the T&E community with a baseline definition and understanding of digital twins as well as related technologies such as digital models and digital shadows. We recommend continuing collaboration among government, industry, and academia to provide further guidance on the value of digital twins across program lifecycles. T&E Lessons learned, best practices, and standards must be developed and shared across the T&E and acquisition communities to advance the application of digital twins. As we look for ways to accelerate the delivery of capability to our warfighters while also achieving deeper understanding of systems’ operational performance, now is the time to start developing and using digital twins in T&E as well as the other phases of acquisition.

Thank you ITEA Hampton Roads Chapter for hosting the workshop and the many companies and Universities who participated.

1. AIAA Digital Engineering Integration Committee. (2020). “Digital Twin: Definition & Value—An AIAA and AIA Position Paper.” AIAA: Reston, VA, USA.

2. A. Fuller, Z. Fan, C. Day, and C. Barlow, “Digital twin: Enabling technologies, challenges and open research,” IEEE Access, vol. 8, pp. 108952-108971, 2020.

3. A. Rasheed, O. San, and T. Kvamsdal, “Digital twin: Values, challenges and enablers from a modeling perspective,” IEEE Access, vol. 8, pp. 21980-22012, 2020.

4. E. VanDerHorn and S. Mahadevan, “Digital Twin: Generalization, characterization and implementation,” Decision Support Systems, vol. 145, p. 113524, 2021

5. A. Tolk, P. Barry, and S. C. Doskey, “Using modeling and simulation and artificial intelligence to improve complex adaptive systems engineering,” International Journal of Modeling, Simulation, and Scientific Computing, vol. 13, no. 2, pp. 2241004.1-19, 2022.

6. Bharathy, C.S.S. (Feb. 2021). “The Need for Digital-Twins in Smart Manufacturing.” BW Businessworld.

7. Cortes, L.A., Wong, M.K., Cortes-Morales, A., Wells, D., & Daly, J. (Nov. 2021). “Advance M&S In Acquisition T&E.” McLean, VA: MITRE.

8. Office of the Deputy Assistant Secretary of Defense for Systems Engineering. (June 2018) “Digital Engineering Strategy.” https://man.fas.org/eprint/digeng-2018.pdf [Accessed February 2024]

9. Semeraro, C., Lezoche, M., Panetto, H., & Dassisti, M. (Sept. 2021). “Digital Twin Paradigm: A Systematic Literature Review.” Computers in Industry, vol. 130, p. 103469, DOI:10.1016/j.compind.2021.103469.

10. Tao, F., Sui, F., Liu, A., Qi, Q., Zhang, M., Song, B., Guo, Z., Lu, S.C.Y., & Nee, A.Y. (2019). “Digital Twin-Driven Product Design Framework.” International Journal of Production Research, vol. 57, no. 12, pp. 3935–3953, DOI:10.1080/00207543.2018.1443229.

11. Grieves, M., & Vickers, J. (2016) “Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems.” In: Kahlen, J., Flumerfelt, S., Alves, A. (eds) Transdisciplinary Perspectives on Complex Systems, pp. 85–113, Springer, DOI:10.1007/978-3-319-38756-74.

12. (6 Dec. 2022). “Digital Engineering and Modeling Practices.” Department of Defense Handbook, MIL-HDBK-539. https://quicksearch.dla.mil/qsDocDetails.aspx?ident_number=285031 Accessed February 2024

13. Kritzinger, W., Karner, M., Traar, G., Henjes, J., & Sihn, W. (2018). “Digital Twin in manufacturing: A categorical literature review and classification.” Ifac-PapersOnline, 51(11), 1016-1022. DOI: 10.1016/j.ifacol.2018.08.474

14. Barricelli, B. R., Casiraghi, E., & Fogli, D. (2019). A survey on digital twin: Definitions, characteristics, applications, and design implications. IEEE access, 7, 167653-167671.

15. Hu, W., Zhang, T., Deng, X., Liu, Z., & Tan, J. (2021). Digital twin: A state-of-the-art review of its enabling technologies, applications and challenges. Journal of Intelligent Manufacturing and Special Equipment, 2(1), 1-34.

16. Jones, D., Snider, C., Nassehi, A., Yon, J., & Hicks, B. (2020). Characterising the Digital Twin: A systematic literature review. CIRP journal of manufacturing science and technology, 29, 36-52.

17. Amazon, ‘What is Digital Twin Technology’, https://aws.amazon.com/what-is/digital-twin/, accessed 15 March 2024.

18. Digital Twin Consortium, ‘The Definition Of A Digital Twin’, https://www.digitaltwinconsortium.org/hot-topics/the-definition-of-a-digital-twin/, accessed 15 March 2024.

19. IBM, ‘What is a digital twin?’, https://www.ibm.com/topics/what-is-a-digital-twin, accessed 15 March 2024

20. Nividia, ‘What is a Digital Twin’, https://blogs.nvidia.com/blog/what-is-a-digital-twin/, accessed 15 March 2024.

21. Unity Technologies, ‘What is a Digital Twin?, https://unity.com/solutions/digital-twins, accessed 15 March 2024.

22. Wikipedia, ‘Digital Twin’, https://en.wikipedia.org/wiki/Digital_twin, accessed 15 March 2024.