Digital twins are understood and applied in a wide variety of ways. Across sectors and disciplines, interpretations vary, some emphasising technical architecture, others focusing on data flows, decision-making, or long-term system change. This diversity reflects the richness of the field, but it can also challenge efforts to build shared foundations.
In 2023, the National Digital Twin Programme (NDTP) undertook a wide-reaching engagement exercise to better understand how digital twins are perceived, defined, and used in practice. This process brought together views from academia, industry, testbeds, standards bodies, voluntary institutions, and multiple government organisations.
Through more than 40 in-depth interviews, a key insight emerged – while definitions of digital twins differ, expectations for them tend to cluster in two broad areas.
The first relates to programme-level principles or guiding values that shape how NDTP operates day to day. These include commitments to openness, value delivery, and inclusivity. They guide our ways of working and ensure that the programme remains accountable to its purpose and stakeholders. These principles are outlined in full on our Principles page.
The second relates to expectations for how digital twins themselves should be developed, regardless of context, sector, or scale. These expectations are more technical or design-oriented in nature, covering issues like security, adaptability, and ethical use. While these ideas appeared frequently across existing frameworks and organisational principles, they were often blended or inconsistently expressed, creating ambiguity about intent.
In response, we developed a dedicated set of Digital Twin Development Attributes which are clear, community-informed expectations for what digital twins should be able to demonstrate in practice. These attributes are not sector-specific or prescriptive, but they reflect shared aspirations about what “good” looks like in the design, deployment, and operation of digital twins.
These attributes are also intended to support more consistent procurement, specification, and evaluation processes. By articulating a clear set of expectations, they help buyers and delivery teams align on what is being requested, developed, and assessed. In this way, the attributes serve not only as a reflection of community input, but also as a practical reference point for those commissioning or supplying digital twin solutions.
Grounded in the voices of our engagement community, they are intended to support coherence across diverse efforts, acting as a reference point for organisations developing, procuring, or integrating digital twins in real-world contexts.
Looking for a technical definition?
For those seeking a more technical and formal understanding of what constitutes a digital twin, including how it differs from digital models, shadows, simulations, and control systems, we have developed a comprehensive Digital Twin Definition under our Resources section. This resource is particularly aimed at those involved in procurement, delivery, or integration activities, where clarity of scope and capability is essential.
Digital twins, individually and when connected, should have the following Attributes:
- 1
Safe
Capable of maintaining the required state of relative freedom from threat or harm caused by random, unintentional acts or events. - 2
Secure
Capable of maintaining the required state of relative freedom from threat or harm caused by deliberate, unwanted, hostile or malicious acts, delivering the appropriate and proportionate levels of confidentiality, integrity, and availability. - 3
Trustworthy
Able to provide sufficient confidence that the system is behaving as intended, with appropriate measures around transparency, accessibility, and control. Transparency of provenance and quality of the data and/or models will allow users to determine fitness for purpose in relation to their decision-making requirements. - 4
Ethical
Adhere to, and promote, individual rights, privacy, non-discrimination, and non-manipulation. - 5
Sustainable
Are affordable economically, socially and environmentally in terms of the resource consumed, at the point of development and over time. - 6
Adaptable
Are robust and resilient – able to perform under both ordinary and unusual conditions, as well transform, renew and recover in a timely way in response to adverse events. These two aspects work in tandem to deliver reliability.
Extendable in use – allowing for extension, customisation, and flexible modularity, to accommodate diverse user needs and applications.
Extant – can be maintained, updated and supported throughout their lifecycle the ensure they are up-to-date, relevant, and available. - 7
Interoperable
Able to be connected and integrated with various systems, technologies, and platforms, enabling exchange of data and information across without loss of quality.