A National Underground Asset Register: Delivering Value with Flexibility

[Tammy Au]

Join us for the next video in our series on Tuesday. Neil Brammal and the CDBB team will host a live chat session at 10.30. Bring your questions.

 

 

[Tom Hughes]

Welcome to the start of today’s Digital Twin Talk on A National Underground Asset Register; delivering value with flexibility and a big thank you to @Neil Brammall from the Geospatial Commission for joining us.

We’re looking forward to your thoughts and questions related to Neil’s talk – and maybe posing one or two of our own.

As with all our Twin Talks Neil is online from 10:30am to 11:30am to answer your questions. To join the discussion please add your thoughts by using the “reply to this topic" box below.

[Tom Hughes]

To start the discussion off - Your talk provides details of the two pilot locations. I understand that the areas were selected to allow the pilot to build on existing work in these areas. My assumption would be that there may be some differences between the very dense urban environment in London boroughs to a wider region of the North West of England. Did the different locations result in any different findings that will be applied to NUAR going forward?  

[Neil Brammall]

Good morning. As @Tom Hughes said, I'm here online and ready to answer any questions.

[Neil Brammall]

Hi Tom.

Regarding the pilot locations, it's correct that a key reason was the ability to build on existing work, but the differences between the areas held great appeal as you say. Not only were we able to see differences between dense urban networks and sparser suburban and rural locations, but there were also real differences (and some similarities) between how data is represented in the two areas. These differences allowed us to build our understanding of how to visualise and interact with data in different locations, but also an understanding of the flexibility that we would need to provide for such a wide range of asset owners.

[DRossiter87]

Hi @Neil Brammall, I'm asking while I watch (apologies if it is covered mid-vdieo).  Does NUAR include the capture information related to any associated easements and if so, how is it captured?

[Neil Brammall]

Hi @DRossiter87. At this point in time we haven't included Wayleaves and Easements in the pilot system. We have done some more generic work around representing "Zones of Interest" etc though, and this is an area that we continue to work on and refine. We're very much aware that just the "centreline" of a linear asset isn't always the full story from a safe dig point of view! What's your view on representing wayleaves and easements specifically?

[Koye]

Hi Neil, thank you for your presentation in regards to the pilot how did you manage integration of data sets from different suppliers? 

Also going forward with NUAR, will there be a requirement for users to have an ordnance survey licence to access data?

[Guest]

Thanks @Neil Brammall for the introduction. I have just had a look at the report mentioned in the video, via "National Underground Asset Register project update" https://www.gov.uk/government/news/national-underground-asset-register-project-update .  To bring so many stakeholders and data together with modular data intergration and ensure interoperability, technology is one of the key enablers, although not as crucial as the mechanism you build trust, engage stakeholders, and focus on tengible use case- eg. "safe digging". 

I am interested to understand more about the technical implementation of the OGC- MUDDI [Model for Underground Data Definition and Integraion]  A ref I found online: https://docs.ogc.org/per/17-090r1.html 

1. A quote from the OGC Engineering report- one of the Design Requirement "The model should derive from or map to/from existing candidate models such as INSPIRE (Infrastructure for Spatial Information in the European Community), CityGML UN ADE (Itility Networks Application Domain Extension), GeoSciML..."  which existing models have you used, that are common for the UK undergrand data?

2. When you standardise data, did you use relational database or graph database?

3.  Are there any open and shareable parts of the NUAR? eg.  spatial identifersor shared asset.

Edited August 4, 2020 by Steven Zhang
correct typo

[DRossiter87]

1 minute ago, Neil Brammall said:

 What's your view on representing wayleaves and easements specifically?

My experience was mainly with DCWW easements when I worked in local authority in South Wales.  These were given to us as 3m from centreline.  I'm not familiar with other easement/wayleave criteria but I would think that a "Easement/Wayleave Offset" length property; attributed to the system (as opposed to each component) would be practical?

[Tom Hughes]

I think that the use case driven approach you described is very relevant to this community. The decision to focus on a small number of specific high value use cases appears to have played a big factor in the success of the NUAR pilot. Do you have any insights on how you established and prioritised use cases, and are there any recommendations you would give to others doing similar use case orientated work?

[Neil Brammall]

Hi @Koye. The integration of datasets from different suppliers is one of the main challenges of this sort of combined platform. We've worked with dozens of asset owners during the pilots, and there is a real disparity between how data is represented and stored at source (quite understandably - the data held by asset owners needs to support many, many use cases and business processes of course, not just NUAR!)

We're trying to manage this in two main ways:

- by adopting a harmonised data model (based on the evolving OGC MUDDI standard) that allows us to represent disparate source data in a standardised way

- by providing as much flexibility as possible in the ways that asset owners can provide data to the system

There are 700+ asset owners in the UK, so a one size fits all approach to ingesting data won't work. The pilot projects have given us great confidence that, while this is a significant challenge, it is not insoluble.

 

Regards your question about an OS licence, this honestly isn't my area, so I will need to take this offline. Our analysis of data requirements includes consideration of the background mapping of course, and this question also cuts across who will have access of course - initially, access will be to statutory undertakers and their supply chain.

[Neil Brammall]

Hi @Steven Zhang. (I remember meeting you at an ODI workshop some time ago, in the days when people gathered together in the same room!)

To take your questions one by one:

1. There are many different data models implemented by the different asset owners, and very little common ground between them. The challenge has been to map these many heterogeneous representations into the MUDDI conceptual model, and the future challenge is how to scale that. During the pilots we have worked intensively with individual asset owners to understand these mappings, and we have also tried to take a "step up" and identify common ground across different organisations in the same sector. As we move forward we'll be looking to consolidate that common ground within sectors to come up with a standard set of mappings that we will have to make minor adjustments to as new asset owners come on board.

2. For the pilot, the data model has been implemented as a relational database, but early sandbox work did include some work with graph databases.

3. The NUAR data is at this stage is generally deemed to be at the "Group-based Access" point on the ODI data spectrum, i.e. accessible to authorised users only, with some locations and assets deemed to by highly sensitive and subject to greater access control.

 

[Guest Mehran Eskandari]

Hi Neil,

May I please ask about the steps beyond the asset registry phase. Will you consider using sensors? and the interactions between the assets? 

Thanks. 

[Tom Hughes]

Your talk provides a link to a published project update – Here is the link for others in this conversation https://www.gov.uk/government/publications/national-underground-asset-register-project-update/national-underground-asset-register-project-update

Clearly underground assets have broad relevance in the Built Environment and from your project update it looks like you have been able to benefit from developing relationships with other teams working on the same challenges.  

A key objective of the DT Hub is to support collaboration through encouraging collective learning and sharing experiences on digital twins

How did you go about identifying collaborators, what was your experience of collaborating with others working in this space, and what do you feel the DT Hub could do to better support collaboration?

[Neil Brammall]

Hi @Tom Hughes. Regarding identification and prioritisation of use cases, this was driven overwhelmingly by consultation with the stakeholders. It almost pains me to say this as a "Technical Advisor", but stakeholder engagement has been by far the most important part of the project! We have listened to the pain points of a huge range of stakeholders, and identified use cases that are achievable and help to address those pain points. So, everything has been driven by consideration of the problems - so often it's the other way around, with a cool solution being developed then finding a problem for it to solve (I'm also guilty of this, believe me!)

The other key thing is that, having identified our use cases, we've kept a laser focus on them - it's so easy to get sidetracked, try to solve everything and end up solving nothing. 

[Tom Hughes]

An open question to all in the discussion. Has Neil’s presentations provided any key insight into how value and flexibility can be delivered? I would particularly like to hear from any guests (not yet a DT Hub member) who are online, what will you be taking away from the talk today?

[Neil Brammall]

Hi @Mehran Eskandari. There are no plans currently to extend NUAR beyond the initial use cases - that's part of the "laser focus" on use cases that I referred to in my answer to Tom.

That's not to say that the work going on with sensors etc. isn't really exciting though, and our aim is to build the NUAR platform in a flexible way, so interoperability and extensibility are possible in the future, for instance if there's a strong use case around sensor data and - most importantly - a strong desire for our stakeholders to take advantage of it.

[Tom Hughes]

Towards the end of your talk you cover the need for a compelling use case for data consumption to address the challenges associated with outbound API connection to NUAR. Is this an area you could see input from the DT Hub community assisting to define potential use cases for data consumption?

[Neil Brammall]

Hi @Tom Hughes. Regarding collaboration and relationships. As I said, stakeholder engagement has probably been the most important part of the project to date. My colleague @holgerkessler is the stakeholder engagement lead, and he maintains a huge and evolving plan to track who we need to talk to and engage with. Our delivery partners at the London Mayor's Office and Ordnance Survey have also maintained close relationships with the asset owners in the pilot areas.

I don't think there's an easy answer to this - for us it's been about prioritising engagement above almost anything else, and spending lots of time and effort on talking, listening and engaging with anyone and everyone with an interest in this area.

Forums like this are great in providing a focal point, and a place to discuss and share - any way to get a receptive audience in one place is very welcome!

[Neil Brammall]

Regarding outbound APIs @Tom Hughes - absolutely, we're always open to thoughts and suggestions from this community, and would appreciate these insights.

 

[Guest]

22 minutes ago, Neil Brammall said:

Hi @Steven Zhang. (I remember meeting you at an ODI workshop some time ago, in the days when people gathered together in the same room!)

To take your questions one by one:

1. There are many different data models implemented by the different asset owners, and very little common ground between them. The challenge has been to map these many heterogeneous representations into the MUDDI conceptual model, and the future challenge is how to scale that. During the pilots we have worked intensively with individual asset owners to understand these mappings, and we have also tried to take a "step up" and identify common ground across different organisations in the same sector. As we move forward we'll be looking to consolidate that common ground within sectors to come up with a standard set of mappings that we will have to make minor adjustments to as new asset owners come on board.

2. For the pilot, the data model has been implemented as a relational database, but early sandbox work did include some work with graph databases.

3. The NUAR data is at this stage is generally deemed to be at the "Group-based Access" point on the ODI data spectrum, i.e. accessible to authorised users only, with some locations and assets deemed to by highly sensitive and subject to greater access control.

 

Hi @Neil BrammallThanks a lot for answering all my queries! They are very helpful. The reason I asked the third question, was because I was wondering if there might be some common sharable identifiers for various digital twin, so that digital twins can talk to each other. Currently, the unique identifiers for Property and Street are made open, if they are referenced in NUAR, probably, somehting worth considering (if it is safe to do so). ? 

Yes, we met in a workshop at ODI (in April 2019) before the "Sharing engineering data for the public good" event, https://theodi.org/project/sharing-engineering-data-for-the-public-good/ . It is very inspiring to see NUAR have made such good progress! Looking forward to see the full operation national NUAR.

Edited August 4, 2020 by Steven Zhang

[Neil Brammall]

Thanks @DRossiter87 for the comment on wayleaves and easements. Yes, we're certainly looking at representing "Zones of Interest", which could encompass rights of access, into the data model.

[Tom Hughes]

Members of the DT Hub community may not be aware that the NUAR project has been shortlisted for the 2020 Digital Leaders 100 list.  The DL100 list is created by public vote and celebrates the teams and individuals who are working hard behind the scenes to secure the UK’s Digital Transformation.

You are able to vote for NUAR until noon on 28 August 2020.

[Neil Brammall]

Thanks @Steven Zhang. Yes, you make a very important point about the need for persistent and unique identifiers to support interoperability. There's quite a wide variety in how assets are ID'd at source, and how we manage this is currently very high on my list.