# TEA-DT Workshop (Infrastructure)


:::success
⏱️ **Agenda**

**Times:** 13:00–16:50 (UK Time)
**Breaks:** we will take breaks every hour, but please feel free to step away from the computer whenever you need.

1. Introduction to project and workshops
2. Introductions from participants
3. Structured discussion around assurance needs and capabilities
4. Presentation 1: What is TEA?
5. Wrap-Up
:::

:::info
ℹ️ **About this Document**

Please use this document to answer each of the following questions. You can provide your answers anonymously. To do so, please ensure you are not signed in, and do not write anything that may reveal identifying information.
:::

## Questions

### 1) What does ‘assurance’ mean to you?

- DT twin prototype -- will people on the market ==trust== it or not?
- Trustworthiness is one thing. ==Validation== is important too, to *prove* the trust.
- Validation means a lot of different things to different people in different fields: researchers may look at ==observations/graphs==, but for safety-critical DTs, it might be more involved than that. ==Qualitative validation== too -- talking to pilots and ATCOs for example.
- Assurance of sensors in autonomous transport systems = how do we know that they're telling the truth, with certainty? How do you know a meter is a meter and a second a second?
- Confidence in something turning up at the simplest level. Peace of mind... Safety... ==Single source of truth==, in relation to data or personal information.
- ==Standards== — arrived to through standards bodies and a collaborative process. In DTs, when building out new business models, ideas, technologies, where there are no standards, they form as a result of people speaking about them a lot. Some rigid standards in autonomous vehicles area...
- ==Negative safety== — what would happen when something went wrong? Better assurance leads to better blame attribution, responsibility attribution and compensation liability. Might also contribute to assurance of the technology.

### 2) What processes of assurance do you currently follow? (and Why?)

- Explainability. The input data I know. When I check the result, I need to be able to explain what the model has done with the input data (in relation to a physical principle, for example).
- Why is it important? It follows the same principle as equations explaining the natural world. How do I continue that in a data-driven research world?
- Peer-review is obviously important.
- Ongoing process of communication with pilots and ATCOs.
- Shadow-trial (where ATCOs were able to play around with the DT + rated their experiences)
- Following various ISO standards = formal governance process.
-

### 3) Are there any guiding principles that you follow?

- (==Explainability==, see above)
- Environmentally sustainable/==sustainability== = basic ethical principles (EU) -- beneficient, non-maleficience, etc... Global environmental justice.
- ==Fairness==. What is at stake here? Decision that is directed towards something/someone. Here, we need principles/theories of justice more generally.
- ==Safety==. (Heavy vehicles carrying 80 people on dangerous roads.)
- Involved ==internal approval process== (everyone on all levels need to be aware of solution architecture designs).

### 4) What techniques or mechanisms do you use to assure these principles? [or gaps, where you have trouble assuring principles?]

- Formal methods for ==explainability== (see presentation).
- ==Interoperability==. Mostly have worked on validation... I'd imagine with other DTs, you'd want to think about an API that would connect relevant data. Simple details that come from an airspace DT to an airport DT. Think about what an AI looks like. Then, challenges with validation come in again.
- Important to prioritise interoperability early on, in creating DTs in different teams. CReDo (Climate Resilience Demonstrator) is a good project that provides examples of how we should build interoperability.
- Not just API development/technical task. it is also business resource sharing, task transfer/integration.

---

## Links shared in chat

- From my NPL colleague Louise Wright - NPL Head of Digital Metrology - Hi Chris. So far, what we have done can best be described as "think pieces". [Digital twins for metrology; metrology for digital twins (iop.org)](https://iopscience.iop.org/article/10.1088/1361-6501/ad2050/pdf) and [How to tell the difference between a model and a digital twin | Advanced Modeling and Simulation in Engineering Sciences | Full Text (springeropen.com)](https://amses-journal.springeropen.com/articles/10.1186/s40323-020-00147-4) with the second one being written first.
- We're interested in the model updating aspects and uncertainty quantification associated with the model outputs in particular.
- [National Timing Centre programme (NTC)](https://www.npl.co.uk/ntc)
- [NPL National Timing Centre innovation nodes](https://www.npl.co.uk/ntc/innovation-nodes)
- [JET Connectivity Launches Floating 5G for Offshore Wind Farm](https://jet-eng.co.uk/jet-connectivity-launches-floating-5g-for-offshore-wind-farm/)
- [UKRI Digital Twinning NetworkPlus: DTNet+](https://www.dtnetplus.ac.uk/)
- [DSIT's Introduction to AI assurance](https://assets.publishing.service.gov.uk/media/65ccf508c96cf3000c6a37a1/Introduction_to_AI_Assurance.pdf)
- [A pro-innovation approach to AI regulation: government response](https://www.gov.uk/government/consultations/ai-regulation-a-pro-innovation-approach-policy-proposals/outcome/a-pro-innovation-approach-to-ai-regulation-government-response)
- [Advancing Data Justice](https://advancingdatajustice.org/)
- [AI Ethics and Governance in Practice: AI Fairness in Practice](https://www.turing.ac.uk/news/publications/ai-ethics-and-governance-practice-ai-fairness-practice)
- NPL/NMI collateral on Assurance and standards incl. globally: i) [CIPM MRA - BIPM](https://www.bipm.org/en/cipm-mra/) and ii) [Data science - NPL research areas](https://www.npl.co.uk/data-science) and case studies. Happy to connect as needed.

## Feedback

### 1. Do you have any feedback about the TEA platform?

- Answer 1
- Answer 2
- Answer 3

### 2. Do you have any feedback about the TEA-DT workshops?

- Answer 1
- Answer 2
- Answer 3

### 3. Do you have any feedback about the TEA-DT project, more generally?

- Answer 1
- Answer 2
- Answer 3