Navisworks for Small Projects

ADAM WARD: So welcome to this class. This class is Navisworks on small(ish) projects. The introductions I'm Adam Ward. I'm director at Spaced Group.

I'll explain who Space Group are in a second. And I'm joined by Paul from our Space Architecture business. Firstly, I apologize. I'm from Newcastle, so I've got a bit of an accent. And every time I present at AU, everyone thinks I'm Scottish. And some people struggle to understand us, because I can get a bit fast.

So I'm going to try to keep it slowed down a little bit. But if you think I'm bad, wait till Paul gets up, because see, he is Scottish, and he's going to really confuse us all. So I head our BIM technologies teams, and I'll explain who they are in a second-- BIM star teams and our R&D team as well.

And Paul basically manages our architecture business. He's an architect. Before we begin, you're probably wondering why I'm here, why Paul's here, because this isn't our presentation. In fact, I don't even know the guy who put this presentation in. I don't know if he's in the room. I don't think he is.

So I'll explain why we're here. So I was sitting at home one day basically in the shower. And what happened was I got a phone call from one of my ex colleagues who I used to work with who's now at Autodesk.

So I picked up the phone, and on the other side was James. And James said basically someone had dropped out last minute last week, and he needed somebody to fill the class. And he didn't want to cancel it, and he needed somebody. So he asked me, and I went-- yeah.

So basically, it needs to be a class on small projects. Now, and of course-- now this works as my thing, so not a problem. And he was like awesome. And I was like, don't worry.

But then he went-- well actually, it's next Wednesday. And I thought, oh, hell. So I basically had two days-- well, we've had two days total to pull this presentation together. That's James. Yeah.

So this is how I felt a couple of days ago. Not sure where to take this.

[LAUGHTER]

So I got to the office. And I decided to put Paul into it to give him a bit of-- and Paul was kind. Can't do it. Too busy.

Then I said it was in Vegas. And Paul was yeah, I've got it. So you just decided to come.

So literally we got to work, and we did the presentation. So apologies for no handouts being available. That's the reason why. We weren't supposed to be presenting here. We've come in last minute to present the class which we've taken over, for want of a better word.

So Space Group-- in our group of businesses, we've got an architectural business, which is a 60-year-old architectural business, fully digital. Everything we do is in Revit. We've got BIM technologies, which is a BIM consultancy in the UK-- works on huge projects, which are going against this presentation. We'll come onto that in a second.

We've got BIM star, which is a content business-- builds content and hosts digital content. And we've got our R&D business, which are a bunch of clever developers who work across the group and make round pegs fit into small holes-- square holes. Like I said, in BIM technologies, we work on huge stuff.

So when I got asked to do this presentation, I thought it wouldn't be a problem. Then I looked at the projects we were working on. And everything's-- I couldn't find anything going to $75 million.

So obviously, when you're coordination on huge projects like shopping centers like this, which are really complex. It's critical. Chelsea Barracks, a 2.5 billion pound project-- it's a no brainer. You're going to do it, because it's going to save you money. Skip that one.

Facebook's new UK headquarters-- there's over 600 separate subcontractor models in this model. We have one issue. So you've got IFC. You've got Tekla. You've got Revit, and you've got all sorts of really exotic stuff, and it's all 100% coordinated without any issues, which is great, huge projects. So clients can spend this money to get this level of reduced risk, basically.

Another good example we're working on at home is the Brent Cross Shopping Center. It's like 12 buildings in one. It's a 2.5 billion pound project, and they're rediverting the motorway, rediverting the river.

It's an absolute monster. It's still at conceptual design stage. But these are the models, so Civil 3D. You've got Revit. You've got everything in there at a quite high level.

So this is the kind of stuff typically we worked on. But then again, we also use the same technology for small buildings. So in our architecture business, where Paul works-- Paul delivers a lot of supermarkets for a company called Lidl. And he's going to present later a case study on that, how we use digital technology, Revit, and Navisworks to actually manage that process.

Same for Babbitt's. Babbitts is the biggest housebuilder in the UK-- do 17,000 houses a year, and they're fully digital, so everything is delivered from a build information model. So when people say BIM's only applicable to huge projects, that's not actually true. People are doing it now. You just don't hear about these smaller projects.

So what I'm going to try to do in this presentation is try to destroy some myths. So one of the myths is Navisworks is only useful on large, complex projects. And that's completely wrong, because we've got quite a good workflow to take a Navisworks workflow, what works on huge buildings-- it also works on small buildings.

And once that workflow is understood, actually, there's not much difference, whether you've got a $500 million pound building or a 500,000 pounds supermarket. The workflow and the way you do it, it's the same. And I'll show you some of that, because I'm going to be getting into the software as well.

Another myth is modeling in Revit and Navisworks takes long and does not bring value. So a good example is Babbitt Homes, they build 17,000 houses a year. They probably built that house there 5,000 times over the last five years. So they coordinated it in the real world.

Everything that's wrong is now engineered out of that house. So what benefit does Digital, and Revit, and BIM bring to a company like Babbitt? So this presentation's going to be split down into the following sections. I'm going to explain a little bit about what Navisworks is, and how it fits into our workflow, and how we use it.

How many people use Navisworks, out of curiosity? Awesome. So some of the stuff I might be showing you, you might already know. Some stuff, you might pick up. And then I'm going to be jumping into some tips and tricks.

Basically I'm going to show you the way we do it on every project and show you how we can do it really quickly and easily. I'll be about 20 minutes. Then what Paul's going to do is Paul's going to jump in and do a case study on Lidl, which is a supermarket, a much smaller building.

I'll show you another project which we're doing. It's a little island just up the north of Scotland-- tiny little island, the MOD. Think I'm allowed to show it, but just don't say anything.

Then I'll show you Park View Student Accommodation, which is a modular housing block for students-- again, small project. That's about 10 [? million. ?] So why we use Navisworks?

We use Navisworks, because we found it's the only tool that allows us to do these core features in one environment. So for example, check into things and clash detection-- there's lots of tools that do that. I would argue that Navisworks does it quite well. It's probably one of the best.

Again, there's lots of tools. There's 4D build sequencing. Navisworks isn't the best. Probably Synchro is the best, if you want really good stuff.

There's quantity take-off in there. I'll explain how we use, I think, the second. And most importantly, it's really useful, as a viewer.

So when you have these models which are made up of lots of different models from different formats, Navisworks is a good way to federate all them together and actually view that information. The reason we use Navisworks is like I said, there's a lot better tools to do some of these things. But what Navisworks does is put it all in a singular environment.

So the same model you are bringing to [INAUDIBLE] to do your class detection, at the same time, you can extract all the quantities from the same time you can update your 4D build sequencing. So it's a central hope of everything we do in BIM technologies. We do use other software, and it varies on projects.

We use Solibri. We use custom tools and all. But generally, this is our Navisworks stuck. And we're trying to get rid of paper. So in design team review meetings, we have big touch screens.

And the model comes first. So we're actually use the model as a way-- where in the old days, you used to put paper on the desk and talk about an issue, we're now doing it in a collaborative environment. And again, like I said, you've got 4D logistics sequencing, but also mobile technology, because this technology is now getting more accessible.

You can chuck an NWD up into BIM 360 or A360, even let somebody view that in an iPad really simply. So Navisworks is our primary coordination platform for construction stage. We do use Autodesk BIM 360. But we use Navisworks for the construction aspect, because currently, BIM 360, in my view, you get down into the level of detail you need to to actually coordinate to that level.

And like I said earlier, it's used on every project, regardless of size. So what I'm going to walk through now is actually some tips and tricks on how we do it on every project. Initially when we get a project, it takes us half a data set it up in Navisworks. And I'll explain what that involves and what we do.

But actually ongoing-- so every fortnight, when we get new models from the teams and bring them all in, we can rerun that report, reissue the quantity take off, update everything, and issue a report within 45 minutes typically, regardless of the project size. And I'll show you how we do that. We have created our own cloud platform. We did that, because the report [INAUDIBLE] Navisworks is a bit dull. So I'll give you a little look at that as well.

So this is the typical process, regardless of building size. So we've got lots of different models from lots of different people. As I'm sure, if you all use Navisworks, you understand that.

What it allows us to do is bring in logistics models. So we don't just model the architectural structure and the MEP. We model the tower cranes. We model the house fencing. We model temporary works.

We link in the program by [INAUDIBLE] power project, if it's applicable. We don't do this on every project. But if it's applicable, we always add time. Because in a building, not everything's built at once. So there's no point doing clash detection against architecture and MEP if some of that MEP-- or sorry, structure's is a better example-- if some of that structure is temporary.

So by adding time to the equation, you're making it a bit more advanced when you're doing your quote and issue detection. I describe it as if you digging a hole in the corner of the site on day one, and you're digging another hole a meter away on day one, you've got a temporary collapse issue. Move those two activities apart, problem gone. So a lot of that.

In the UK, this normally happens after stage three, four. Then when the subcontractors get involved, we start getting models from them. Now obviously, there's a lot of work we do before this about wrecking the BIM execution plan, getting the standards in place. There might be EP about actually what formats, what data's in the model, but I'm not going to cover any of that. All I'm going to try to explain here is we get lots of formats.

And in the UK, IFC is a normal format. So that's normally where the subcontractor models come in. And that's because some of those subcontractor models is created in really exotic pieces of software on some of our projects-- stuff like tier and stuff like that for facades, for example.

And during the design stage, it's quite common to use BIM 360. So even though I said I'm not a big fan of it myself yet-- the new version is awesome, but I'm not a big fan of the current version. The next generation version is good.

It's a really cool tool for real time coordination during the design stages. So the architect, the engineers should be still coordinating amongst themselves before [INAUDIBLE] give [INAUDIBLE] information, mind you, who's normal use, to manage that. So that's quite a good platform for that. But we use Navisworks as the central hub, for want of a better word.

And like I said earlier, it's this idea of it being the central platform for this single source of truth model. So the workflow is we start off with three models. We bring them into a federated model. I'm going to go through all this in a second.

Navisworks will go and find all those issues. But we don't just leave it there, because if you just go and clash structure against MEP, you're going to get 10,000 clashes. And 90% of them are the same.

So for example, a pipe going up through a tower-- it's going to hit every slab on the way up. But actually, in my view, that's one issue. Move the pipe over, and the issue's gone.

So what we do is we group those individual clashes into real issues. So we take those 10,000 issues and make them into 50-- sorry, 10,000 clashes and make them into 50 issues. And I'll show you how we do that really quickly instead of sitting down, doing it manually.

And then what we do is we stick it in our wallboard platform, and we produce really nice visual reports and a dashboard of the project as well. So we find the individual clashes. We find the 10,000 clashes. Those are reviewed.

We've actually automated this. I'll show you this in a second. And then they turned into issues.

But even then, the software, Navisworks, isn't going to go and find all issues. Because a column in the middle of a room isn't a clash. A door in the wrong position isn't a clash, but it's an issue.

So our information, mind you, is still scanned over the model, looking for these issues visually. And I'm going to show you how we do that as well very quickly. Yeah. federated model, coordinated review.

So the first thing is when you first start the process, you need to take the time to set up the models to work. And it's worth spending a bit of time doing this. So what we do is we already create a master home view.

So what you'll see on there is two views at the top, which are our master views. Home view is basically asymmetric, all the real world pullers. A coordination view is we've turned all the individual models to be their own distinct color.

So for example, architecture's always gray. Structure's always yellow. MEP's always pink. Facades are always green, and there's lots of [INAUDIBLE] and all.

So what it means is when you're looking at the model, you can instantly see what belongs to who. And we just use color overrides to do that. And I'll explain why we use color overrides rather than the visual profile.

Then you create your default search sets, which I'll come onto in a second. And always delete any folders which comes through default from Revit or whatever platform you're coming through, because they will slow your model down massively, unless you need them. And turn up your resolution of your Viewport Exports in your developer settings, which I'll show you and all when we get into it.

So that'll just make the quality output look a lot better. So you can see on the top up here, we've got the views. I've got my home view. If I click on the Home view, it always defaults back to that view, real world colors.

And I know this isn't a 10 million pound project. Then if you click on Coordination-- sorry-- if you click on each one of the floor plans, it slides them down. So that's our default setup when we set a project.

Obviously, you can very quickly jump to a level, et cetera. But what we do is we set up what's called a coordination view. So if we click on the coordination view now, it adds all those colors I've just mentioned. So all the architecture is great. The structure is yellow.

In this one, the health care is green, the health care specialist. Now what that means is you can zoom, and you can instantly see stuff that's wrong. Now you'll notice what happened was I'm still using the same views.

But by clicking on the master view before I jump into the overview, it's defaulting-- basically remembering the master view when you go into the sub views. It's a little tip that when I train people in Navisworks, who's been using Navisworks for years, you just miss it for some reason. So you can see that there-- I can instantly see that pipe's in the wrong position.

So now if I click back on Home View, then click on one of the put views, you can see it went back to real world colors. So the way to do that is when you're creating these home views, you right click on it, which I'll show you in a second. Go to Edit, and make sure that those two boxes are ticked. And then update the view.

Then it's going to remember that view. So when you click on that view first than any other view, it always remembers that view setting on there. And it helps us visually-- so when we're talking about views and Navisworks on small projects, sometimes you don't need to do clash detection. You can just use that process and instantly see stuff that's wrong.

So for example, there's a column here. And you can see the structural column's yellow, and the architecture is grey. So you can instantly spot stuff which just looks wrong. For example, there's a bit of MEP floating there.

So by just doing that color trick, it makes things a lot easier. Now, I know you can use a [INAUDIBLE] profiler to do that, which is the proper way of doing it. But we don't do that.

The reason being is there's another good benefit you can get by not using it. So if you just override the colors manually, what it does is it means next time you get your new model linked in, it'll actually show you what's changed in the model. So you turn everything pink. You bring in a new model, a UMEP model.

And what happens is you can instantly see everything which isn't pink-- it's stuff which is new, what stuff which has changed. So you can very quickly see areas where you might want to be looking into. And it's a two second job just to click on the model and just click turn back to pink-- override the colors back to pink. And you can see on there, on structure, we could see the structure engineer, you must've been messing around with the cause, because they're back to normal color and not yellow. So it's a very good way to visually see where the teams have been working.

The second little tip is keep clash sets to a minimum. So this is a bad way of doing it, MEP versus-- this is how most people I've seen do it-- MEP versus Structure MEP, versus Architecture MEP, versus Facade Structure, versus Architecture. And you have a lot of clash sets.

What we do is we go for a layered approach, so there's about three or four in every project. So what we do is first thing we clash MEP against structure. Then what we do is we've already clashed MEP and structure. So then what we do is we clash MEP and structure against architecture.

Then we clash MEP structure and architecture against facade, so it's instantly [INAUDIBLE] six or seven clash sets and reduce them down to three. And that works really well. The other tip-- and I'm going to jump in the software and show some of this stuff in a second.

So the other tip is search sets-- always use search sets. Keep them simple. So if you're working primarily with Revit models, what you don't want to do is just load in an MEP model, a structure model, then clash your MEP against your structure, because you're going to get all sorts of problems like you're going to get analytical lines. You're going to get stuff which is voids, which you don't want.

So what we do is create search sets for the basic core elements. And normally, it's just the Revit categories. We do have architecture, structure, and MEP. And in architecture, we have a primary, which is doors, windows, curtain walls, stairs, handrails-- but it ignores any walls, any ceilings, any floors, because everything hits them.

And we have a separate set for those. So you can see on here is a typical example. The other tip is what we do is we spit out concrete slabs and walls from the structure models. So when we're creating sets for our structural model, we create a structural frame and set. And we create a structural concrete set, for want of a better word. And that allows us to very quickly-- I'll show you. It's easy to show you.

What it allows you to do is very easy. It highlights the issues in really complex areas behind our 10,000 clashes. So you can see on there. So you can see.

I've got clash set 1, clash set 1 0. So clash set 1 is MEP versus structure. Well, actually, I've got my MEP versus structure split into subcategories in old steel and concrete.

The other little tip is take time to create your clash views. So what we don't do is issue that, because if you give that to a contractor, the first thing he's going to say is what is that red thing, and what's that green thing? So we tend to create nice views. And I'll show you how we do that as well.

And again, just another example of that's what we go for, rather than the normal default way of doing it. Well the little tip is turn graphics up to 11. So when you're exporting these things, we always turn on Screen-space Ambient Occlusion.

Now if you're working on small projects under 10 million, you could leave that on all the time, and you'll be fine. And it you can see, it put shadows on-- it highlights everything so much better. So one of the first things we always do before we print any report is just turn that on. 90% of the time, we leave it turned on. And it's such a huge project.

The other one is you notice when you export your reports from Navisworks, the images are tiny. And they're pixillated when they expand. So there's a secret hidden setting to make that go away.

So if you go into Options and you hold Shift while you press Options, you get hidden features. You get developer options. And a lot of the things in here-- you can break things, if you touch it.

But the one that changes if you click on Viewport Defaults, you can change the image height, and the image width, and the anti-aliasing as well. So generally, this is like 400 default with zero in there. So that's our standard.

It takes a bit long to export the views, but not that much longer. But to get that, you have to go into special settings. Or the way you do that is you go up to your whole shift while you press Options.

And when you press Shift, you'll get this special-- you'll see there's a lot more stuff in there than you normally get. You get stuff for-- basically can change default, how stuff works. There's a lot of techie stuff in there. But if you click on Viewport Defaults-- I'm going to find it.

That's wrong. You can see in Clash Detective, you can change the colors where red and green-- you can do all that stuff. But if you go into Interface Viewport Defaults, you'll see there's now a new-- I don't know what I'm doing on this video. You'll see there's a new bit there.

You've got press Shift to get that. And now, whenever you export your report to Navisworks, it's going to come out a lot better quality. So that's the small images on the bottom. If you click that, it'll be pixilated.

If you do that little trick, it'll will not be pixilated and look quite cool. But even then, the "out of box," this sucks, so we don't use this. So the problem with it is the images are tiny. The NWF file based workflow for collaboration is not ideal. So chucking an NWF file around the different people for comment-- it's an old fashioned way of doing it, chucking a file around.

It's not very interactive-- no historical context on issue. So if I wanted to say, well, this report doesn't tell you much. It says there's 100 issues. But on some projects, 100 issues might be good. On some projects, it might be bad.

So there's no historical context about what that means. And of these tools like BIM 360, which are moving towards that dashboard workflow-- so what we do is we custom built our own tool to do it. And we've been using it for years, called Warboard.

And you just upload to Navisworks XML or BCF, if you're using Solibri. And it gives you a dashboard over time. So you can see if the number of issues is going up, going down, how much data is in the model. You can then dive in and look at the individual issues, comment on them.

So this allows anyone who doesn't have a copy of Navisworks-- even the client visually going to comment on it, et cetera, et cetera. It's fully cloud based. What happened last year is everyone who we worked with on projects wanted it, so we made a commercial. So you can sign up for a free trial, if you want to go on it.

You can put who's going to resolve by and get an e-mail saying resolve this, et cetera, et cetera. And it prints off really nice reports. And like your [INAUDIBLE] and Navisworks reports, you're going to get full blown-- good-- You're going to get a status page. You're going to get full images, comments on, et cetera.

So one of the big things I mentioned, one the biggest tips I can give you is to use Groups. Now I don't expect, if you've got 10,000 clashes, to actually be able to go and group all them manually, because that's going to take you hours or even days. So I'll show you a little trick on how you can do that very quickly.

So again, you can see. In here, you've got three or four clashes. You've got a clash here with two elements. You've got two elements here, you're clashing. You've got this clashing.

So in this view here, you've probably got about six or seven clashes. Well actually, that one view is enough to say that's the problem. It's not coordinated in this area.

And the same for slabs. So the reason we split our slabs off on our clash sets is at early stage of the project, you might not be putting the holes in the slabs. So what we do is we actually issue each one of these for each floor level. Now because it's a Navisworks group, it still remembers all those individual issues. And it won't resolve that until they've all been resolved.

So it still works. It's a better, easier way to show that. And I'll show you how we do it. So I'm quickly going to go into clash detective. And I normally have double screens, so this is going to be a bit crazy.

And what I'm going to do is I'm going to go and just turn off architecture and just have structure versus MEP, because that's what I'm going to go and do a clash on. I just zoom out, so we see what's happening. Now I know this isn't a 10 million pound project. But if I can do it on this this quickly, it will show you how you can do it on any project.

So first we'll do MEP versus architecture. So what I'm going to do is going to add a new set. And I'm just going to call this MEP versus-- [INAUDIBLE] structure. And I'll steal. That'll do.

And I'm just going to put demo in front of it. Then what I'm going to do for that is I'll just pick me WSP, which is my MEP. And I'm going to clash it against my structural steel frame.

So basically it's going to ignore any concrete and just clash it against steel. So when I click the tolerances, I'm not going to go into any of this. I'm assuming this is all familiar.

So I'm going to click Run. That's going to comeback with hundreds of issues. In reality, that's not that many. Normally, you have 10,000 or something in there.

So the way we very quickly understand what's going on and turn those 200 issues into 10 issues is this. So the first thing we do is we create a single group. And I'm just going to-- I'll just call it something ridiculous like that, so I can see it.

If we now click Highlight Over, it actually highlights all the issues. But obviously, you wouldn't want to issue that in the report. You'd want to be a bit more specific.

So for example, what I would do on here is I would maybe go find an issue where you can see there's one, two, three clashes there. Navisworks will want to do all those individually. So the trick is if you now select one of the items which is hitting everything, and then you click inclusive on here, that's going to find all the clashes which apply.

Then if you click on them and group them, and then call it 001. So that's pulled then out of the main group. [INAUDIBLE]. Then if we go back to here now-- so then what we do is now we walk around here looking for clusters of issues. So there's a nice cluster of something going off.

So what I would do here again is pick this, this, this, this, be a bit more specific. And then what I would do is I would pick a piece of it here. I would actually-- Inclusive. It'll go and pull out all those issues. I need a double monitor.

So I've instantly grouped all those issues. And then I would go and do that until there's nothing left. Then whatever is typically left at the end-- and this will take half an hour. What's left at the end, I would normally just put as kind of reviewed. It's been resolved, for want of a better word.

So then what I would do is I would click on each one of these. And then generally-- that doesn't tell me anything. So what I would do is I always go down and below, and do that. Still not really telling as much, but trying to get a view like that.

Then I would go on this one, find out what the issue is here, try to get a view. That shows that. That'll do. Obviously go and do all the revisions and stuff.

So very quickly, what you're doing is you bring in your 10,000 issues into one big cluster. You're looking for clusters of issues. And then you're pulling those individual groups out.

And very quickly, you can produce a report. The really cool thing is in Navisworks, even though that's classed as one group, or until all those individual subelements are resolved as well, this will always remain active. So it's quite a good way of doing it.

Then what we would do is it would export to XML, upload to our board, and it would go and do all your fancy reporting, et cetera. So let's quickly jump back in here. That's what I've just shown.

On some projects that we model ourselves-- bear in mind our information, mind you, so we don't always model all the models. But on some of the architecture ones, we do. We do go sometimes a bit above and beyond. But stuff like modeling things like angles on valves or door swings can make a difference, clashing with light fittings, and stuff like that.

So on some projects, especially smaller ones, you can push the limits a bit. On this project, it's Google's new headquarters in London. We model and risk. So you can see on there, we've got an element here with the key issues assigned to it, which can be then tagged off on elevations, on plan, et cetera.

But interesting. You can clash against that. So when that comes into Navisworks, you can clash risk against activities. Stuff like tower cranes-- we can model swings on tower cranes and add that into the mix as well. So it's not just clashing elements. It's clashing bits, and pieces, and all.

So what I'm going to do now is pass it over to Paul to go through a case study of Lidl.

PAUL BROADFOOT: Good morning. So you probably know Lidl. They're a discount retailer. We've worked with them for a few years.

And the whole point of this earth-- I've been involved with BIM for over 12 years. And it's how using BIM, and Navisworks, and having full clash detection and a coordinated model makes a difference to projects. And these are typically 3 million pounds. So you'll get an idea of how much detail we put into them at that point.

But why should Lidl care? At the end of the day, for every day, it generally always comes down to money. Being able to do things faster, quicker, more efficiently is in their interests, and it's in their interests. And when you've got an expansion program of 1.5 billion over three years in the UK alone, you get an idea of where those efficiencies and improvements in every store type or every store and region or region, you actually start to see the full benefit across the whole batch.

One of the other key things is time. When you're one of Lidl's main competitors-- certainly in the UK, it's Audi. And one of the key things is how quickly they can actually get to site and get sites bought, get sites built, because the faster you're sailing, the faster you've got an edge on your competitors.

There's just a couple of indications-- 200 stores per year just in the UK alone. Another key thing for the likes of Lidl is brand. It's got to be delivered in a consistent way. It's got to be.

So when you go to a store-- and from time to time, you get the same store stype. So what's the problem for Lidl? You've got a head office in Germany. There's your concept store to all over the world.

And then that then is taken by architects throughout the world. And then their store types, you can get inconsistencies. You have issues with quality, and cost can vary from state to state.

So this is where we came in. Basically what we did is we worked 11 stores for Lidl, and then we spoke to head office. And he says what we want to do is we want to do things definitely.

So we're going to be using BIM. We're going to be seeing how we can actually use coordination and actually develop a design so that we can actually make it more efficient for us. Basically what that means is in that top tier, we get a concept store from Germany. Then the head office level, we then produce 60 to 70 drones which are issued to all these architects throughout the country, sometimes even to Europe. And that includes [INAUDIBLE] quantities, specification, and the Revit models.

But watch. The key thing is that it's data rich models. So their implementation for BIM at the moment is we've produced several store types.

These are issued on a monthly basis. And [INAUDIBLE] all the architects to get those key aspects-- consistency, quality, speed, and cost reduction, because the key thing to remember, even when you're doing projects of a lower value is one error repeated 200 times is 200 errors. And when you do that in every store type, you can see the issues that you have.

So even when you have lower value projects, if you repeat a mistake, it's a costly exercise for any manufacturer, any contractor. No site's the same. So by having the building blocks so that consistency can be created and every region is a massive value to Lidl, because it means you create the components and the framework so that the key things can be replicated from site to site.

But one of the key things is drawing accuracy. When we do a few models, they'll all be fully coordinated. Now at that point, which I should have mentioned is when Adam talks about Google [INAUDIBLE] and you get 400 BIM models, we're actually only talking about five subcontractor models.

So you'll have MEP, structure, landscape, and the architectural model. Sometimes you'll have facade models. But in projects of this size, scaled four to five actual models come to get that Federated model. It's typical.

And I don't know whether you have a similar experience. But we view that that's enough detail to make it efficient and to keep the process moving quickly. One of the key things is, as I've mentioned, is drawing accuracy.

All these remedies are taken directly out of the model so that when we actually go to site, that image on the left-- so obviously, that picture's from site. Thus, the cones-- the one on the right was an image directly from the Revit model where the same view-- and that was just an image straight from Revit or Enscape. You use a bit of both when we're having the coordination reviews.

But you get an idea that you shouldn't be able to-- you should. You have the quality, consistency from store to store, because you can check it. You can go on site. You can go on your iPod, and you can actually make sure what's being delivered is what was designed.

One of the things I've been doing recently is we've been actually using virtual reality in walk throughs with the head office so that when we come to little things like where certain signs should be, we have crash protection to the entrance. So we actually go through with the Xbox controller, and we just use that rather than go on a site. So when we talk about efficiencies and design improvements, we're using the model to work on what we're going to be doing next.

We don't need to go to site to see what's failing, because we can actually walk through the space ourselves. Other key aspects is that we can do design options very quickly. If you want to change a tile color, you can actually take a visual straight from the module very accurately and quickly so that you can make decisions.

Different store types, different options, things like augmented reality and [? popup. ?] Sometimes things like augmented reality seem [INAUDIBLE]. But when it's actually forming your decisions, how you're going to design in the future, it's not [INAUDIBLE]. It's actually a design tool that we find quite useful.

One recent thing they've been developing is a bill of quantities straight from the model. Now that is a difficult task. And when you use a product like Lidl-- and I repeat stores-- it gives you the chance to actually achieve that, because you're doing it over and over again, whereas if you're doing a one off bone, it's quite difficult to do a bill of quantities direct from a BIM model, as you may have not known.

What's key is that within that design model is all the [INAUDIBLE] manufacturer components are included if they can be, if they're known. But when you've got a repeat building likes of Lidl, likes of Barracks, when you get store types or house tapes, and it's the repeat building, it's the same [INAUDIBLE]. Make it accurate. Make it exact, because you're taking the quantities and costs, and you can make it so clever that the data is in all of these design models.

What we do is we set up a site for them, so that they can download. So say an architect-- it's every month when we issue all these design models, they can download the Revit model. Now of course, every site's different. But you start to get an idea of the amount of data that's in those models.

So they've got a great template so that they can then amend and change that for their own site specific need. And I keep harping back to consistency, quality, speed, and cost. This ensures that the architect that's delivering the project though on site achieves all of those goals.

Another key aspect is that within the model, you have the downloadable component. So when Lidl or any other clients be used, they want a specific toilet fan. That's what's installed in a model. So when it comes through the bill of quantities, the references and quantities actually state the model that should be installed.

Now that gets difficult and [INAUDIBLE] one off projects. But for Lidl, it's very achievable. And that means that on a site by site basis, it'll be the same toilet effect used in every single store. And we achieve that by distribution of the correct data within the model, even right down to the correct toilet fountain itself.

This data actually-- it's all [INAUDIBLE] compliant. But what we're doing is we're storing up our infrastructure for Lidl so that that can then be developed for the asset management and years to come for them as well and how they want it developed moving forward. That was a whistle stop tour of Lidl, same with Barracks. And it was just to give you an idea of just one such project of less than 5 [INAUDIBLE], which we do on a day to day basis-- which you've got any questions, feel free to come up after the session, and I'll try and help where I can.

ADAM WARD: [INAUDIBLE] . I'm running out of time, so I'm going to quickly jump on to St. Kilda's. So this is the MOD facility. So obviously, BIM's great on huge projects. It's great on repeatable projects where you want to get the risk out of there.

But we also use it on small one-off projects. So this is an MOD facility. I can't say what it does, but it's accommodation and an energy center for something. And it's in the middle of nowhere. So this is on a tiny-- you can actually walk from one end to the island to the other in less than 20 minutes.

So that's the building, if you can see it hidden down there. It's in a conservation zone, et cetera. And this is where it's at. So it's in the UK. If you go to the north of the UK, it's a tiny little island there.

And it's an R&D facility for-- you know what it's for. I don't think anybody knows what it's for, but it's a little building in the middle of there, and it generates power for something. That's our little island there.

So on this building, it's tiny. So that's the accommodation block. It's where the people, the researchers, the scientists are going to live when they're out there, because this place is in the middle of nowhere. You can only get there by a helicopter.