Part Three: Digital twin collaboration

Wayne: If you're building a true digital twin, you want the fidelity of that twin to be as close to the fidelity of the real world as possible. That's the whole notion of twinning.

Disclaimer: This podcast is brought to you by the team at Boustead Geospatial productions, in conjunction with our affiliates - Esri Australia, Esri Singapore, Esri Malaysia, and Esri Indonesia. To get your hands on more short, sharp and immediately usable resources, head to the GIS Directions Podcast website, and check out the show notes.

Wayne: Welcome to GIS Directions. I'm Wayne Lee-Archer.

Ta: I'm Ta Taneka

Mary: And I'm Mary Murphy.

Wayne: And welcome back, everyone, for the third of our three-part series on digital twins. We've dived into the topic over the last two episodes and discovered it's quite a complex discussion.

Mary: So, let's do a quick recap of where we are at.

Ta: We've covered the first two pillars that enable a digital twin. So that's data capture and integration and real-time and visualisation. I know those are the first two, we have two more to go.

Mary: So, I feel good that we've built some of it, but have we actually thought of everything? Typically, we're building a digital twin with that audience, our stakeholders in mind, and we want to engage them, right? We want to communicate with them, we want to show them some outcomes or products or results, etc.

But we also might need to think at this stage about where and how we can get those results or those products. Where do they come from? And they come from the analytics, the analysis. So when we're building that digital twin, we should not only plan for how it will be shared, but also how we should use it for analysis as part of that workflow.

Ta: You're absolutely right Mary, we can start to see the living digital twin in effect, in action and today we'll briefly touch on sharing and collaboration and analysis and predicting, and then we're gonna dive into the technical ‘how to’ of building a digital twin, that’s my favourite part, I cannot wait.

Wayne: But first of all, let's briefly touch on that sharing and collaborating and analysing and predicting. It's a really important step and it really sets the foundation for our future preparedness for different scenarios.

Mary: Yeah, and it's one of those pillars if we're starting with that particular pillar, which is the one I love the most, shall we say, we're talking about that ability to be able to simulate, evaluate, and then potentially predict, that’s usually when we're talking analytics, that's what we're gonna do.

But we can't do any of that. We can't analyse without the data, we can't analyse without the correct data, good data, and we can't analyse without an understanding of the methods we might want to leverage or use. So if we start to build that digital twin without considering, as an integral part of that construction process, our data and the methods we want to use or are going to use, or techniques we're going to apply, then we may not actually be able to even leverage that digital twin as expected.

So that's why I think it needs to be part of that construction conversation and processes, it always comes back to the audience and your purpose, so the analytical capabilities are going to differ according to those needs from one digital twin to another. So, I suppose the next thing there is how do we consider analytics in that construction process? So, we can use things like the spatial analysis workflow is a good place to start, ask, explore, analyse, model, interpret, repeat, modify, present, and make decisions then ultimately.

Another one that I always point people through, just think about those six basic types of analysis, right? So what have we? Proximity, overlay, that big bucket – the statistical ones, your temporal analysis, your network and utility things, and then your 3D. You're probably gonna wanna work with a combination of those things, and they're not gonna happen by magic or by accident.

Wayne: Yeah, it's absolutely a combination of all those things. And there are lots of technical words there, Mary, you know lots of GIS, there are real-world ways that we're actually seeing this kind of thing already realised in very, sort of, simple and useful ways.

I'll call out the product SmarterWX, which is actually allowing, you know, multiple utility organisations and governments and council organisations to share and collaborate their information together to actually get some real action you know. There are example cases where there might be a roads team who are gonna come and pull up a road and then a paving team that is gonna lay it down.

And then an electricity network that might be laying some more electricity assets within the system. And they're all doing that at different times. And what you find is you get these repetitions of tasks, the redo of tasks and inefficiencies in the system. Once people come together, once they collaborate and share, and once they actually then do their analysis, both in time and space, in a product like, you know, SmarterWX, we can actually reduce those costs and reduce those inefficiencies.

So, you know, we're seeing real-world cases of these that wouldn't typically be counted as a digital twin, but they still rely on these same pillars of twinning.

Mary: And that's a combination of the sharing and the analytics, just overlapping so much that they're one and the same. Another one that I came across actually when I was having a look at some ways people are using digital twins, obviously sharing the digital twin, but obviously then sharing the results of their analysis is from Incheon Metropolitan City in South Korea.

So they're using that for disease modelling and prediction. And I think they're aiming to use it for operating a flood prediction and monitoring system. Those things don't happen by accident. They had to think about the types of analysis they would like to conduct and build it as part of the process and then share it in whatever way they need to with the relevant people.

Ta: Okay, I promised you guys some treats, we’ve got treats in our bags – five key steps to building your digital twin, from the proverbial and virtual ground up. The very first step, super important – get rich data. The richer the data, the better it is and bring it into ArcGIS Pro. Ready, Steady, Pro!

Wayne: I could not agree more. When we were talking before about the visualise and the other pillars, we didn't really focus on the data itself. If you're building a true digital twin, you want the fidelity of that twin to be as close to the fidelity of the real world as possible. That's the whole notion of twinning.

You want, real-time feeds where you can, you want overlays from all sorts of different organisations. You go to the BOM and get that weather data in there.

Traffic data, bring in the high-fidelity data, reach out to your partners. Reach out to people you haven't reached out to in the past. Bring that data in and get it into Pro. Because that's gonna be your starting point for all of this.  

Ta: Right, that's where we're building the groundwork. So open a template or an existing project and add your data by connecting to your organisation's database or publicly available data from Living Access of the world or from ArcGIS Online.

So we’ve got our data, we've got a rich set of data, and then step two, look, we are cartographers by nature and at heart. So visualise your data and make your data work for you. Select some really great symbols to represent your data. Set your reference scales for better context to your area of interest.

Wayne: Don't go overboard with that symbology, though is my real hot tip here. Yes, we want to get that high fidelity data in there, but you've also gotta remember that we are cartographers, and we should be doing things in a, a visually smart way to represent information. Don't forget about things like zoom scales. Don't forget about turning things on and off where it's appropriate because I can tell you now, you are going to put a massive load on the rendering engine if you are gonna try and render every single streetlight in your country when you're zoomed out at the country level, it makes no sense. It doesn't really provide that fidelity, it's just more blobs on the map. So be cartographers.

Mary: Yeah. Don't just put it there because you can and you have it. It becomes unusable, we don't want that. We want people to go, wow, we have all this data and we can use it, and it's efficient and it's effective, right. Get smart about how we have symbols turn on and off where we need them to as we navigate in and around saying what you’re labelling.

So symbology, very important, and we can get things into Pro, we can get those old style files as well from ArcMap and import them into ArcGIS Pro. I'm on the same page as you there Wayne, we definitely need to be smart about our cartography.

Ta: This is really important for governing bodies, like in public safety and emergency management, and for accessibility, you know, ArcGIS Pro has some great new functionality around accessibility.

For example, anyone who's colourblind. So yeah, bring in the right symbology for your audiences so that you can symbolise and make a really, really available map for everyone.

Mary: But sub-setting your data is a good one as well. So clipping it using extract by mask, depending what you're doing.

Data engineering workflows are excellent, so use the data engineering workflows within ArcGIS Pro, for example, to suit your analysis needs, and then even using those reference scales. So getting in there and doing that so that we're only showing what we need to show when we need to show it.

Wayne: Yeah, look, absolutely. I could not agree more with everything that you've said there, Mary, I can finally let the cat outta the bag, I've worked on the largest utility network project that has ever been undertaken in GIS so far, and that's for the Energy Queensland Corporation in Queensland.

Here we're talking about 195 million pits, pipes, power lines, and poles, try saying that fast, all managed in one GIS. And what I can tell you is smart visualisation is absolutely the key. You don't want to try and render every power line poll in the state when you are zoomed out at the state level. It doesn't make sense. It doesn't add any quality to the visualisation and data.

So just use your smart defaults, use your zoom and scale levels, use your areas of interest, clip extract by mask all of the tools that you've just mentioned there. Absolutely essential.

Mary: It comes back to that same thing with the real-time data. Do we need all the data all the time? We don't need all the features all the time.

Ta: Okay, so step three is the most exciting part for me, enabling your 3D data with the quick touch of a button in ArcGIS Pro. A lot of capital works, building information models, utility data. Wayne, shout out, they all have a height, or a depth value assigned, you know, so it's known as the Zed value. So go ahead and enable it in ArcGIS Pro.

Quick and easy to do this on the ribbon, click the ‘view’ tab in the view group. Click ‘convert’ and choose to convert to a local scene. Then on the ‘appearance’ tab in the extrusion group, click the ‘type’ button and then choose ‘maximum hype’, and then view your masterpiece. You know, everything is extruded to the height of the data that you have.

Wayne: It sounds just like four or five clicks, Ta. It can't be that easy.

Ta: Absolutely. 1, 2, 3, you're there. Look, an added step you can actually enable shade and shimmer on your 3D enabled data before you publish your scene to your web GIS. In scenes illumination works by simulating sunlight and shading on the faces of your 3D features, such as your extruded polygons or your buildings.

So quickly to illuminate a scene on the view tab, again in the scene group, this time, click the top portion of the illumination split button. And then, you know, you can use the split slider to change the time of day so you can see what the sun looks like, you know, at midday, at nine in the morning, at three in the afternoon.

And, for additional benefit, you can access all illumination settings from the map properties under the illumination tab. Then, once you're good to go with all your shade and shimmer, publish your scene to your web GIS.

Wayne: Super important for digital twins because, it all starts with the map, now in this case, we extend that to, it all starts with the map and the scene. So we wanna be building a great map, and then we wanna build a great scene out of that as well, because we want the 2D and the 3D that's really, you know, the key about this whole twinning thing.

Mary: Step four. Okay. So we've got that scene in our organisation's portal, so we want to be able to share this in some way, shape, or form. Again, there's no point in doing all this amazing work if we're not gonna tell someone about it, and we wanna tell someone about it in a way that's gonna be useful for them.

So, we might want to use Experience Builder, some other applications like a dashboard, for example. So there's lots of ways that you can tackle this. But from within the actual scene that you've created, you could click that create app button and then you can choose from the options that you get there.

Things like instant apps, Experience Builder, StoryMaps, I did see as well there’s a 360 VR experience in there now. You just do that from within the scene, or you can launch those apps individually as well from the app launcher in ArcGIS Online, for example.

Wayne: It really is about the shared scene and map, isn't it?

Mary: Once we get it into our portal. Yeah, that's huge.

Wayne: I wanted to have a big shout out to Experience Builder here first and foremost. Experience Builder lets you actually join those two and 3D views together, which is super important.

You can have a map on one side, you can have a 3D scene being viewed on the other side. You move the map around and it moves that 3D scene as well. And so I think, you know, that's really cool and a great way to be actually be visualising our digital twin.

Mary: Especially now that we can do that in Pro, because we can now work with 2D and 3D and Pro, it's nice that that's actually carrying through to some of the things we can create and push beyond our desktop.

Ta: Absolutely. That quick click of a button, remember, convert from a 2D map to a 3D scene, a local scene, and then yeah, world’s your oyster.

Wayne: Alright, step five guys, and this is my language here. I would encourage users to develop and gamify and enhance and collaborate. Digital twins provide a really excellent way to collaborate in two directions. Now, what I mean by two directions, this is not just collaborating out to the people that you wanna share this with.

This is actually collaborating inwards as well. You don't expect to have all of the data that you want. We talked about data fidelity back in some of the earlier steps. You can't expect to have all of that data yourself. And so what we're seeing with this whole motion towards digital twins, is we're seeing new partnerships and new collaborations form. You're seeing roads and transport authorities sharing their data with Bureau of Met.

We're seeing all of these different organisations that probably would never have crossed paths in the past, coming together and actually cross sharing their data. So don't forget that two-way collaboration and sharing. Don't forget the people who might be able to feed the data into your system.

But of course, the next step is about sharing as well, and so how do we then take our twin, which is all based upon the map and the scene, how do we then go and turn that into something useful? Obviously, we share it on portal. We share it up on, ah, ArcGIS Enterprise. We can build some cool mini apps using Web AppBuilder or Experience Builder, which are great tools in there.

I do have another hot tip for everyone, and these are the ArcGIS solution templates, the add in for ArcGIS Pro. Now this gets forgotten a lot. We'll throw a link to it in the show notes, but it's an add-in for ArcGIS Pro that actually allows you to take your data and spit out specific applications for your industry sector.

So, if you are in the construction industry, you can do things such as site inspection or site selection applications based upon your maps and your scene. You can do that kind of shadow cast analysis, so you know you're building a new construction in a new area. The little old lady down the street wants to know that her washing's gonna dry because it's going to be, you know, within sunshine during the day and not shaded by this giant building that you're gonna be putting there.

So you can pump out those kind of applications right there in ArcGIS Pro. It's a wizard-based sort of approach. Asks you a couple of questions, you check a couple of boxes, fill in a bit of, of information, and it will spit out and publish into your portal, these specific applications for your industry sector.

So that's my big call out here is, start using the tools that are there and the templates that are right there at your fingertips. ArcGIS Web AppBuilder, ArcGIS Experience Builder, got great templates built in. And then we can use those solution templates from the solution templates add in as well.

Ta: I love what you've just said there, Wayne, because we can use this rich information to start our predictive analysis, it provides this great foundation for us to then share and cross collaborate with our key audience, whether it's within organisation or peer to peer, or just with the general community. I’m, I’m all about everyone, right. How can we all contribute back to the society that we live in right now?

Okay, that's great you guys. We've got five steps to digital twin success and future preparedness in different scenarios, like population growth, infrastructure planning, and application of sustainability measures.

Okay, so what's this tech stack for our five steps? Mary, I know that you're jumping up and down.

Mary: Wayne just, you know, he took the rug from under me there with some of the big ones. So well done you. Obviously we need to talk about the usual suspects, right? So ArcGIS Pro’s gonna help you with all of those pillars.

We've got data, we've got visualisation, we've got analytics, and we've got sharing capabilities. So ticks all of the boxes there. We can work in our 2D and our 3D, which is great. And then when you're starting to think about your sharing and collaboration, we're looking at your portals, right?

So ArcGIS Online, your enterprise portal. And then things like ArcGIS Image or ArcGIS Velocity and, might help you with that.

Wayne: Don't forget velocity for that real-time data ingestion from your sensors out in the field, okay?

Mary: Right, the real-time data allows you to crunch those numbers. And then we have, portal for ArcGIS, obviously enough, your Web AppBuilder, Experience Builder. We also then have some little offshoot things that people don't think about that might help you with the analytics, especially side of things. So ArcGIS Insights was a good one. ArcGIS knowledge, relatively new.

ArcGIS notebooks. If you wanna jump into the world of scripting, etc. maybe do some machine learning and Ai, yeah why not? So there's lots there for you to play with as well.

Ta: And my personal favorite, and Wayne and I have worked on this quite a bit, machine learning and Ai, automate everything that you can when you can, and let your machines and your tech work with you and for you as well.

Wayne: Rise of the machines. I am embracing our new machine overlords.

Mary:  That's it.

Wayne: Let's hear about some real-world success examples of Digital twins. I know Mary and Ta, you've got some hidden away there that you've got his favorites. What have we got?

Mary: Well I already mentioned the Incheon one earlier on again, around the disease modelling and prediction. Definitely worth going reading up about that. The Grenadian in one is pretty good as well. So they've worked with, there's a whole heap of Grenadian officials worked with analysts from Esri to deploy, I think it's artificial intelligence capabilities was the main kind of starting point with that.  

So we were using deep learning models, for example, to identify buildings. So I think within one day the analysts were able to extract 55,000 built structures,

Ta: Oh wow.

Mary: Right. And there's lots to that as well. Again, same thing, we'll pop some links in and some great videos talking about the building of it and the context and so on as well. Really interesting. It's really, really good. It's definitely worth looking up and seeing what they've done, how they've done it, and why they did it.

So you can see the thought that went into the building of the process as well, which is, again, I'm complete nerd about that. So I like seeing people plan. The other one is Pompeii, so a site, I think it's I14, is the name of the site, and they're using a digital twin for data collection, monitoring, progress, sharing, exploratory analysis.

And there's a couple of great quotes in one of the articles that we’ll pop in the show notes as well around the benefits of efficiency and accuracy that they wrote. So having all the documentation, including those photos and drawings available quickly in scaled 3D models and in an ArcGIS Online web scene, has enabled their researchers to disseminate their project findings with everyone from students and GIS lay users to project stakeholders and the public.

I mean that just sums it up. So there are two that I have found on top of the one from South Korea. What about yourself, Ta?

Ta: I just have one really good example that stood out for me, and it's over in the Northern Hemisphere. So Vodafone in the UK, they've created a 3D digital twin of their mobile mast network enabling engineers to make instant decisions to improve services.

So, Vodafone mapped more than 40 million environmental features, including buildings and hills and valleys and trees. All using the ArcGIS Enterprise stack. So this enabled a spatial component to huge amounts of data and it gives a 360 degree view, virtually of the network, on any device over a secure connection.

I thought that was wild, like 40 million. I thought that was amazing. One final metric. Get this, 500,000 network features such as antennas, along with billions of rows of network performance data can be visually presented, and this really results in untapped levels of location intelligence.

Mary: Again, doesn't happen by accident, right? A lot of planning.  A lot of blood, sweat and tears went into that, and that's good on them.

Wayne: So far we've talked about, you know, those applications that you can come and build out of these digital twins, but there's also the little child in me that wants to put on the VR goggles and run around in a real world representation of stuff and shoot up giblets.

So I'm gonna throw out a tool here. Specifically for the dev heads, the propeller heads in the audience. We do have the ArcGIS for Unity Game Engine SDK out there. Once you've built your scenes, once you've built your maps, you can bring in your DEMs, you can actually use this tool to actually build real-world game engine style environments that you can walk through and include in the Unity game app.

Mary, I know you're the training queen. What can we recommend to people to get started in from start to finish?

Mary: Well, as usual, I could go on forever. But let's look at analytics really quickly. So ‘Spatial Analysis using ArcGIS Pro’, the SNAP course. Or we could use the IMAP course, which is the, ‘Image Analysis using ArcGIS Pro’, good ones to start with, if you've gone beyond something like the Essential Workflows. There's also a couple of MOOKs By Esri Inc. The ‘Going Places With Spatial Analysis’ and the ‘Spatial Data Science’ one, amazing, free – get in there.

And with the sharing and collaboration, we've got loads. We've got the ‘Sharing with Enterprise’, so the ESHA course, we've got a dashboard course, we've got a StoryMap course, depends how you wanna share. And then just in general, learning of best practice for analysing and using your digital twin for prediction, for example.

So start to look at things like the spatial analysis workflow. Always a winner to help you figure out and learn what you need to learn to get the digital twin up and running.

Ta: Alrighty guys. Look, I think we've shared some great information in this three-p series on how to use data intake to create a dynamic and living digital representation of our world.

Providing a means for two-way collaboration via a single source of truth that fosters and enables cross organisational sharing and collaboration of data, and gives citizens a voice to share what's happening around them, how it is like it is in real-time.

Wayne: So I think that's about it, folks. We had a mountain of short, sharp, and immediately usable features that will help you to get twinning straightaway.

I'm sure the links in the show notes for this particular episode are gonna be long and distinguished.

Ta: And on that, to help you guys get started with these tips and tricks, we've added all the resources we've spoken about to our website. That's gisdirectionspodcast.com au, including all the information you'll need to get you started with the four pillars of Digital Twins.

We'd also love to hear any tips from you guys, the ArcGIS user community, so please jump onto the website and send them through or connect with us through LinkedIn or Twitter or any of the socials that we're on, and hopefully we can feature some of your tips as part of our resources and future episodes.

Wayne: Stay spatial everybody. Thanks for joining us.

Mary: Until next time.

Ta: Happy mapping.

Disclaimer: The views and opinions expressed in this podcast are solely those of the hosts and do not necessarily represent the views or opinions of the Boustead