Rail Design - Fast and Effective,

QUOC PHAM: I guess we'll probably make a start. Well, thanks again for attending this class and welcome. Thanks to Autodesk for allowing me to present here with Andrew Milford. He's part of Autodesk consulting. But I guess a little introduction about myself. I'm with AECOM. I'm basically a BIM consultant internally within AECOM business in Australia, so what I do is I help, I guess, advise internally as well as externally on our projects.

My wife never really grasps what I do, and in five words I explain to her, I do a bit of engineering, construction, technology, management, and advisory work. That's the best way I can try to explain myself. For those that are here today who's from a civil background side of things? Yeah? And buildings or structures services? Cool.

What our key goals were is to obviously share this information. But part of it was showing you some understanding of the impacts of having good structure, and having some good team structure, definitely helps any process with fast and effective collaboration between multiple products. Obviously in this case Civil 3D and Revit.

Roles and responsibility of design teams, we went dive too much into that but it'll give you some understanding seeing some of the technical demonstrations that we're showing some importance of why we should structure correct roles and responsibilities. Some tips and tricks, obviously, of the Civil 3D tool. Obviously using it in the tunneling workflow perspective, and then obviously diving into a bit of Civil 3D data and how it works in to Revit.

We'll just probably kick off. At AECOM, but I'm sure in your company as well, is we have offices all around the world. But then it's not only just our company but in some projects that, not just us but I'm sure you guys as well, we would have to work with other offices, other resources, other consultants. So not only internally, but externally. This just gives you a demographic understanding of who we may be working with or who you guys might be working with on some of the projects.

This is probably a typical structure that you may see in most projects. You will have a general consortium of project teams. You will have civil, bridges, buildings, master planning, then you have a cost or procurement. These are generally just like a standard structure in probably Australia, and New Zealand. But in the market in the States it may be different. No? Is it very similar? Or would you guys have different, crazy teams?

A part of that may be AECOM but we may sometimes only just work on bridges, and then another consortium may do most of the other works. General structure of the team itself is quite important, because it gives you an idea of who's responsible of what information. So civil, bridges, buildings, obviously have it broken up into architecturals and services. And you can imagine all these teams here will then have to have their certain deliverables, and you can imagine them sharing that information with you, and you vise versa.

It becomes a big multiplication of files going back and forth, especially in a high-pace data sharing environment. You get into another aspect of it now. Obviously in the civil we have to produce rail designs. You can imagine each file may be a certain X amount of DWGs, not to mention DWG drawing files as well. Again, this is just adding to another level of complexity of how much information we need to be sharing.

Obviously, again, with the building side of things, the architect will have his own model. Engineering structures will have their own model. MEP, and you can imagine that's broken up into another three sets of disciplines, mechanical, electrical, hydraulics or plumbing. But then you've got comms and communications. You potentially have securities, tunneling, signaling deliverables as well. Then there's specialties with that.

To give you a context of a particular project we're just using as a scenario today, so I'm based in Brisbane. Andrew here is in Sydney, and I'm sure he'll give you more detail about his background. But for those that I'm not sure where Brisbane is, so Australia's obviously there, but Brisbane is in the east side of Australia. The furthest, probably, one of the-- sorry, the third biggest city in Australia. Being from, I think, Sydney and Melbourne are close around there.

I'm just trying to compare it with other big cities. But in the past Brisbane's always known as a country town, but there's been a new strategy put in place by our councils, and the state government, which is trying to convert into a global city. Cross River Rail is a project that's looking to go ahead, so that's the big dark blue line. The problem with this project is we're waiting for an election to happen in the next couple of months. Depending who gets elected then we get to find out who gets the job. And if I'm still in job, right?

The other strategy that's been put in place is, obviously, the metro line. There's two metro strategies been put in place. This is the first one obviously, and then following through with the second one. But that's basically a further extension from Mater Hill all the way to Mount Gravatt, which is probably the equivalent distance of probably here to the airport of another metro line.

There is existing bus lines, but turning Brisbane into a world city we need to look at probably more sustainable opportunities. Here just highlighting, again, more specifically the scenario we're going through. The Gabba is basically one of probably Brisbane's most iconic stadiums for a game we call cricket. It's not as good as baseball, but it's very similar. Probably, I shouldn't say similar at all, to be honest.

But there's a big, from a master planning perspective, the government is obviously trying to connect communities, as well as people, through to South Bank and then Brisbane. And we have a big development called Queen's Wharf that's commencing as well. There's a lot happening around in Brisbane, so hence why the strategy to put this in place.

To give you a little, I guess, context to the site. So this is the busway, and the route, which generally happens from day-to-day. Obviously during peak hour the buses actually backup right on to-- all the way up to the other side of the bridge. And that's generally between like 5:00 PM, 6:00 PM, and so on. So it's quite a pretty bad place when it basically-- during the peak hour periods of traffic.

So they're looking-- through this master plan, they're looking at what's the options available, in terms of just converting this whole bridge lane into railway. Sorry, as a metro line. And then disconnecting or disusing this road here for vehicle access. Just looking through Google Earth, basically there's a cultural center which is kind of a entertainment area with a museum as well, but that's where the image was before. But following through in here it actually, the busway, kind of cuts into the building through here, just underneath the tracks and through the center. And there's a exhibition or convention center up further north there.

Hopefully that kind of gives you some context of this scenario that we're putting together. These things are not locked in stone, obviously the project itself, but we figured that would be a good opportunity just to demonstrate to you guys some of the best practices and workflows. That's basically an introduction to the project scope, if you will. I'll hand it over now to Andrew to talk further about Civil 3D and how we kick it off.

ANDREW MILFORD: Great. It's working. Thanks, Quoc. Everyone have a good night last night? Few bleary heads here standing in the front. OK, so yeah my name's Andrew Milford. I've been working as a technical consulting with Autodesk for about the last 3 and 1/2 years.

But prior to that I've spent nearly 25 years in the civil infrastructure industry, primarily as a civil design modeler and engineer working for companies such as AECOM, and [? Arcadis, ?] and Jacobs. What I'm going to talk about today is we're just going to run through a typical design process on how we can actually efficiently and quickly model up, and set up of Civil 3D project.

We're going to look at some things to look out for when we're setting up and working with multidisciplinary projects. Is this thing working? Nope. That's OK. The first thing we're looking at, Quoc has already outlined the area of the project that we're looking at. You can see here, here's our proposed rail line heading up across the bridge from Southbank. This is the area we're going to look at actually interacting our Civil 3D model with a Revit model. OK?

Now the first thing we need to start with, there's obviously no detailed survey in this area so we are taking advantage of freely available data. In this case, we have one meter DEM from LIDAR scans, and this is actually freely available from the Queensland government. But you'll find most governments around the world give you access to this sort of information. So whatever region you're in it's certainly worth looking at your local government websites, because they'll have some really good stuff to start with.

A quick shout out to Nearmap here who provided the aerial imagery for this project. You see this number up here, 0.0375 GSD. Stands for ground sample distance, so as you can see, if I get a chance to zoom in here, which I'll have to use the mouse. The quality of the aerial imagery here is absolutely incredible. I could possibly see someone reading his newspaper.

This is a really good starting point to which to begin the project. What we've started with is a ground surface of one meter DEM information, which we've tinned and thrown across into Civil 3D. And we've actually already added it as a data shortcut into our project. For those of you using Civil 3D, is everybody using data shortcuts? Yes. Good.

2017 and '18 brought across Civil 3D corridors, so we're going to be looking at sending across our surface into Civil 3D. We're going to take out rail alignments across. We're going to also throw across corridors, because what we want to do is basically extract solid information from our corridor model and then hook that into our Revit model ultimately.

Just a few points to consider when we're actually setting up our project. Some of the things we obviously need to consider. What other disciplines we're going to be dealing with? And in this world we're never just dealing in a civil space. We have to interact with the structural components, the mechanicals, or any kind of drainage components. And what sort of data types are we going to be handling? Obviously in our case, we're going to be looking at things like LandXML files, we're going to be interacting with Revit files, text files, and things like IFC.

One thing I always like to do with a Civil 3D project, I know it's not essential for using things like Navisworks, but does everybody specify coordinate systems in their projects? Yep. So I guess the main reason for that is, obviously, hooking into GIS data. We deal with a lot of that back in Australia, so that's all located into Civil 3D drawing settings. You find your location, in this case we're in Australia MGA zone 56.

This is probably-- some of the benefits of that, I guess, the coordinate systems also enable us to interact with programs such as InfraWorks. So if we're doing some preliminary concept design work we can actually send the data back and forth between the packages. So that's GIS, programs like InfraWorks, and it also enables you to stream online aerial mapping data.

Custom subassemblies. Now obviously we're dealing with rail. Civil 3D does ship with some pretty-- I'm seeing some heads there nodding in agreement-- some pretty ordinary rail corridor subassembly. What we're going to do is we going to look at some best practice around, let's build out some custom assemblies for our rail.

Now, is anybody here using subassembly composer? Has anybody tried applying cant to a subassembly? Has anybody succeeded in getting it to work? Not many hands, OK. Being at Autodesk I can say that yes, that is being addressed. And it is coming out very soon and I will show you that it does actually work now, which is quite a nice thing. The first thing we're going to look at is we're just going to build a simple rail corridor.

Going to be applying axis of rotation, just a very simple rail just for instances where we don't have anything like sleepers, or track slabs underneath. And actually one thing I will recommend, when you're even attempting to create a subassembly, just plan it out first. OK? I'll either just draw some hand sketches, I'll actually create an AutoCAD drawing based upon that, and I'll consider things such as the point, link, and shape codes that I'm going to be using.

Am I going to enable parameters to let the user add their own point, link, and shape codes? Or am I going to lock them into hard coding? These are the things you need to consider when you're actually setting up your subassemblies. I'd never try and open it up and just start to create something, nothing will work.

We're going to start with some simple rails. Then we're simply just going to do a copy paste, and we're going to start to add ones for track slabs. What we can do is we're going to build this corridor out of really only three subassemblies. The third we're actually going to build a tunnel.

In that case, we're simply just going to copy the basic rail assembly and we're just going to build a tunnel around it. And we can add things like targets, elevation, and offset targets. Now this is just a simple exercise just for the class today. This can be as complex or as simple as you need it to be for your project requirements. Just bear in mind, though, the more points we start to add to our subassembly obviously the longer the corridor regeneration time. There's that fine balance of detail versus practicality.

And lastly we're going to look at an envelope subassembly. Is anybody doing swept corridor envelopes? Cool. I'm saying a few heads nodding there. And this is really just going to help us-- scrolled a little bit too fast there. This one's actually going to help us generate a three dimensional corridor, which we're going to use for clash detection in Navisworks.

I've got a series of videos here, I'm not going live. I've tried that in the past, it hasn't worked very well. We're going to look at creating a simple rail corridor using cant, but we're going to leverage a lot of existing information that comes out of the box of Civil 3D. In this case here all we have is two simple points, a left and a right side point. We have all of our input and output parameters. And all we're doing is connecting in two points and a link.

Notice how we've had this button here before, it's never really worked in the past. You'll be pleased to know now I've set up my left on my right rail is 0.1 and 0.2. I'm just going to put in some arbitrary values. This is good for testing. I'm going to say this a lot in the next couple of minutes. Flex and test your sub assemblies constantly, OK?

And now that we turn the AOR on you can see we're actually getting that axis of rotation working. This is a standard subassembly out of the box in Civil 3D. These things are called sequences. I've copied the sequence from the standard subassembly and I'm just pasting it into the new. I don't want to have to reinvent designing a track rail every single time.

So working in sequences it's basically just like a block of code. You copy and you paste it from subassembly to subassembly. Saves you a lot of time. And you can see there it instantly appears in place. We can do that for the right rail and it appears. And in this case, now we've got the track slab attached as well. And as soon as we connect that up we have a track slab appearing.

The slab itself has been set up with offset targets, as well as custom codes to enable the end user to specify their point codes, as well as the shape.

QUOC PHAM: Andrew, so just on this as well in the general presentation, we tried to upload our presentation and PowerPoint, but we figured that a lot of our stuff is videos. So we're going to try to upload these media files as well, so just probably bear with us after the AU we'll definitely upload all this information for you guys as well. Yeah, sorry to interrupt.

ANDREW MILFORD: For the envelope subassembly we're really just taking-- once again, we're just copying the basic assembly. We're building out instead of creating the actual rail slab we're actually creating two sides of our envelope based upon our design requirements. And we're just creating the envelope to cater for our vehicle throw.

That's obviously the biggest thing we can deal with. A Civil 3D subassembly cannot actually think in advance to cater for the light rail track throw. In this case, what we're doing is we're setting up the assembly with auxiliary points on each corner. We're creating a couple of target parameters in this case, and all we're going to do is-- you can see the targets up here-- we're going to assign the value, which is basically half of the vehicle width, either side.

And in this case, we jump into the sequence there. We're going to select the auxiliary point and we're going to override the hardcoded value with the target. Once again, test it, flex it, make sure everything works. Make sure everything's connected properly. We do the same on the left side. That's a pretty straightforward operation.

And then of course what I want to do is go back to my cant tab in the composer and just test to make sure it actually works properly. Because what we're going to do is we're going to take this, we're going to use Autodesk vehicle tracking a little bit later and generate a swept path. And that'll actually give us our targets with which we're going to create our three dimensional envelope.

Alignment creation in Civil 3D. This is actually-- Civil 3D alignments, I find, are actually very, very easy and flexible to use. You've got all of your standard layout tools, you can import from LandXML, and in this case we're actually simply going to create from an existing polyline. Bear in mind this is just a conceptual design, so all we're going to do is run in. This is an existing alignment I've created just using a standard polyline, I'm pretty sure everyone's created an alignment and Civil 3D before.

In this case we just simply accept the direction, we select OK, and we give it a name. Hit OK and we have an alignment. Notice as I select the alignment we've got that triangular cyan grip indicating IP mode. Notice as I pick that and move it around both of the tangents lose their bearings. I don't like to work that way, I find it actually quite frustrating.

So what we do is break it. And by breaking that we can now actually work at an element level. Notice as we can move one tangent the other one stays locked. That's called the fixed floating and free method. That's a way I always work is I basically detach the elements from the IP mode.

We're just using a standard offset alignment here. Oh sorry, not using an offset alignment, just offsetting using standard AutoCAD commands and just recreating. The reason for that is we need to actually manually come in and increase our radius around the corner. What we need to do now is look at actually preparing our model for exporting into Revit.

What we need to do is we need to have a common point. I need to be able to type my Civil 3D drawing and send it across into Revit. And how do we do that? I need to actually set up-- I'm just going to use a simple alignment label and I'm just going to make the point in the plan where it needs to be. In this case I'm going to use a single station offset.

I'm going to pan to the location where we're going to put our stop. Quoc has already told me what station and offset he needs it to be from the alignment, so a station offset label will simply locate it relative to the alignment with the added advantage, if the alignment moves the base point will move with it. So that when I push the alignment back out to Revit it will actually-- I don't have to change the point at all.

And there's also a single segment point that I also like to use. This is used to just add it along a tangent, or a curve, just to give us a bearing in this case. And this is just used for verification of coordinates and for our bearings. The idea is in Revit what we want to be able to do-- Revit likes to work in an [? ortho ?] space. You've got project north versus true north. Setting up a single segment showing a bearing line across there actually lets us orient our model in Revit.

When it comes to Civil 3D corridors, like I said, we're creating a preliminary corridor here. We're actually interested more in speed and efficiency, because we're going to be looking at multiple design options. Obviously around here you don't want to be creating your models with a frequency of one meter. Otherwise yeah, you've got to get that trade-off between efficiency versus model detail.

In this case, we're simply just going to build out a corridor. It's worth noting this whole process of creating the alignments-- adding a surface, creating alignments, and creating a corridor probably took me about four hours from start to finish. And so what I'm doing now is I'm creating a-- I've just created a single region across the baseline at the start.

I'm now breaking the corridor, I'm splitting the region just using the standard Civil 3D tools. I'm going to break it where we actually need to apply the tunnel. And then all I'm going to do is to select from the ribbon the region properties, highlight the section of tunnel I wish to change, and then simply swap the assembly out. Going from a track slab to a tunnel.

We're going to add a new baseline in this case. I don't want to create multiple corridors, I want it all to be in the one so we're going to add a new baseline.

[ROBOTIC NOISE]

Good morning. And we're just going to add the same slab assembly up to the start of the tunnel and then through the tunnel we're just going to apply the single rails. And you'll see when we can turn into 3D it's actually a pretty efficient way to model. I've obviously extended it a little bit beyond the tunnel. And then all we're going to do is just finish off with the standard rail and slab configuration.

And obviously if you've put in the wrong place just move it around with your grips, OK? It's very efficient and very easy to set up. The balance comes in. There we have our rail going into the tunnel, rail and track. Slab on the outsides.

Updating. One thing I always like to do when I'm setting up a corridor, or any project in general, is I always like to set it up so it's adaptable to change. Is anybody here familiar with using the data shortcuts manager? Yep. Now this is typically a tool that's more or less designed to repair broken data shortcuts, but what it can do is it can actually enable us to do some optioneering really quickly just by re-linking the alignments between versions.

I could have actually just changed the source alignment model if I needed to, say, move the alignment down maybe three or four meters for an option. But in this case I wanted a separate file, so I'm actually going to use two separate files but I only want the one corridor model.

Some of the flexibility-- You can see here this is my source alignments drawing up here in the top. And what I've done is I've just set up a little zoom LISP routine. If I flick to my option you can see there is a definite difference in the two alignments. Over on my data shortcuts I'm going to add my second option to alignments, to the data shortcuts project. And then using the data shortcut manager we're going to come in and we're simply going to swap one out for the other.

Now this has got the advantage that we don't have to actually rebuild our corridor or our regions. This is just automated. I select the object in the drawing on the left, and I map it to the new data shortcut on the right, and I hit that little link button there and you can see it updates the model. Now you got to remember when you do this you need to take the profile with it. It'll actually disassociated.

Same again, MC20, map it to the new MC20 option two. Relink it and it updates. Same again for the vertical relink. You can literally swap out your alignments in a matter of seconds and then all you have to do is come in, rebuild your corridor, you don't have to change anything in it.

Envelope corridors. I'm going to create a separate corridor for the actual light rail swept envelope. In this case, we're going to do this with swept paths using Autodesk vehicle tracking. It's the tool we use commonly in industry throughout Australia for both light rail and for any kind of vehicular turning movements. We're going to, obviously, use vehicle tracking 2D swept path. We're going to build a 3D corridor out of that 2D swept path object.

This is the corridor already built. You can see here, here's our tunnel-- our envelope profile, and you can see the cant applying nicely around the corners there. Finally. In this case we're coming up to our vehicle tracking tools and all I'm going to do is navigate through the 1000 whatever vehicles, and I'm going to select the Melbourne [? five ?] car. Just run through the standard swept path procedure.

And you'll notice now our envelope is actually a little bit wider than that corridor. What we need to do now, once I change the color make it a bit more legible, is I'm going to explode it. Unfortunately, I can't target a block so I have to actually create a polyline out of this. The swept path creates a single polyline, I just break the polyline each side. I'm going to create the left-hand side target, and a right hand target.

All I need to do now is select my corridor. I need to go to the ribbon, at my targets, select my left offset, I'm going to pick the left string that I just created from the swept path, and I'm going do the same on the right-hand side. And you notice inside here the section updates around the corners.

Why we're doing this? We do a lot of this in Sydney. We've got a big light rail project in Sydney at the moment and you can see here-- bring it into Navisworks-- we actually have a lot of heritage trees running down a Boulevard. So they've scanned the entire area, we've taken the scans into Navisworks, created 3D swept envelopes all of it, and the performed clash checks against it for council approvals.

It's a very, very powerful tool. And you can clash it up against anything. Whether it's the Revit stop, whether it's any kind of signaling, any kind of [? architeneries. ?] Anything. Property sets. This is actually a relatively new feature within Civil 3D. It's been around since AutoCAD architecture in 2002, I think. But with 2018 Civil 3D, now when we export solids, gives us the ability to automatically create property set data for use in Navisworks, and for our federated models, and the zone review.

That's pretty easy to extract. That's pretty much all automated. You've got all of your elements, your corridor name, your baseline name, the region name, even the code, the shape code that we've used. But what we're going to look at is how are we going to create our own attributes? And we're going to use some Visual Basic scripting in order to do that, because we can actually take all of these individual elements and start to use VBScript and just do some simple scripting to concatenate and create our own unique tags.

This is the solid model that we've already created from our rail corridor, and you can see when we select it we have our property set Manager with the default, automatic property sets. We're going to add a new one here called Project Information. We're going to use the bodies and 3D solids as the objects to apply to. And we're going to use a formula property definition. Now I'm going to copy and paste because I've done this plenty of times in the past, but the one thing we need to look out for here, I've got this rail attribute ID, is when I paste it in it looks as though it actually has the information from the property definitions here.

You can tell it doesn't work because we've got this gobbledygook up here. What I need to do is actually just highlight across there and actually just double click to insert from down here. And you find that it actually has a light gray highlight underneath it. That means it's actually an attribute of another piece of property data. And so what we can do here is-- all we're doing is just concatenating strings.

We're stripping off the end of the chainages, so we're leaving a simple unique integer, we're applying it here to all of our solid objects. Apply with the button, select, and add. Now when I select it you can see it's picked up the baseline name, shape name, and the chainage. Now we can do this to such a degree-- we've got a set up in Australia where we can actually apply, through just the basic automatic properties, we can apply a unique identifier to every single object within the Civil 3D world. Which then gets pushed into Navisworks and it retains that unique ID.

This is just an example of the property sets that get pushed into Navisworks. You can see there's a unique rail ID coming in here. And, if we were to use BIM 360 Glue, obviously we'd be extracting the same data across the two products. Here's the automatic information and the corridor model info. 2D export to Revit. This is always an interesting exercise. At the moment-- did everyone see the technology keynote the other day? Yeah we are looking to improve it but at the moment we're stuck with what we've got.

The way we typically handle it on our projects in Australia is, this alignment drawing here I wouldn't use my source alignment, I usually create just a separate drawing, data link in the alignments themselves, and add the relevant alignment labels. You can see here we've got the station offset and the single segment indicating the bearing, station offset, plus an easting northing. This is more for verification than anything else. It's always worth double checking when we're importing the model into Revit to make sure that we're in the right spot with our project base point.

Golden rule, no Civil 3D objects. Revit just does not like it. With this drawing I'm going to export the CAD drawing to a simple 2D. I'm going to explode everything inside of it. And the first thing you do, purge everything. Make sure you purge your Civil 3D styles as well. They don't get brought across either, they don't like it. Any xrefs and images just detach it, and always do a zoom extent.

We work out in the real world, those Revit guys work in their own little world.

AUDIENCE: Just boxes.

ANDREW MILFORD: Just boxes, yeah. So do a zoom extent. Quite often you get something that usually hangs down at the origin, just make sure you delete it. Keep the drawing clean, save it off in its own directory, then hand it to the Revit guys for modeling. Yes?

AUDIENCE: If I may ask, presumably you took it off and worked it. Have you guys did a test about the georeferenced coordinates option [INAUDIBLE]. Like when you're collaborating your models in Civil 3D and Revit? Have you tried that? [INAUDIBLE]

ANDREW MILFORD: Sorry what was the georeference?

AUDIENCE: Yeah, essentially we heard about it that in Revit 2018 they added on the option that now you can specify georeferenced coordinates in the Revit models. So I'm just wondering if either company has tried this and test it out.

QUOC PHAM: Yeah, so georeference yes. In terms of if the Civil 3D data that the Civil guys apply has georeference information, and I guess this is the next stage about presentation when I get on, we'll show you how it's as simple as applying the coordinate system and, I guess, making modifications, just slot of modifications. But there's multiple methods to do that. But, yeah it can be done.

ANDREW MILFORD: Yeah. Lastly exporting 3D solids into Revit. Revit by itself can natively read the AutoCAD 3D solids, but the technique I've used, and I'm going to use it here-- Quoc's going to use a different one-- is I'm going to actually export the 3D model to IFC. Typically, I'm sure everyone's seen these sorts of things before, when you export a Civil 3D IFC and try and bring it into Revit, in wall coordinate systems you end up with this. But what we really want is something like this.

To do that we're back in AutoCAD, this is our solid model. I actually went and wrote a small LISP routine here on Civil 3D IFC export. It's about 100 lines of code and what it does is it just enables you to-- you run it, load it, and you basically just select this point here, which is going to be our project base point. Once you select that it basically takes everything, pushes it to the origin, bakes it out to an IFC, and then puts it back. it's a really clean operation, works really well.

And then you can say that it's created the IFC file. I'm going to move that into a separate folder. And then we're going to move into Revit. Now this has already been set up with a project base point. So what I'm going to do now is simply linking the IFC file. And obviously don't get a chance to put it origin to origin, or anything like that. It's going to import and you can see it actually comes in perfectly. It's a really clean operation.

But, the IFC file, we can't do anything with it. We can't push coordinate back, the shared coordinates. What you'll find is when you import the IFC it actually creates a Revit file by default, right? What I do now is detach the IFC I'm going to reattach the new Revit file. Then make sure it's origin to origin. Once again it's in the same spot, it's the same thing, but now we've got the option to publish the coordinates back to that Revit file.

The advantage of that is that when we update the solid model, and we need to push it back into Revit again, all I need to do is open up a brand new Revit project, doesn't matter what template I'm using, import the IFC. It'll just regenerate the new Revit file as long as it has the same name, and then close it down again. And when I go back into my Revit model it'll just immediately update so it's completely automated. With that, I'd like to hand it over to Quoc and he will finish off with the Revit part of the presentation.

QUOC PHAM: Cool, thanks.

ANDREW MILFORD: Doesn't work anyway.

QUOC PHAM: Going back into Revit now. Obviously, once the civil information's been developed. Going through my side of things is the project setup. Commonly time is of the essence. Making sure we have the correct file. You saw before, as Andrew was working through his design, there was a lot of files. So making sure naming conventions is very important. Having the location in the correct position, as well as a location where you know that can be always continuously been update. That's always a good step.

The project team structure is quite important, because you always want to have the same way of doing things, as well as having at least a core team knowing the correct way of doing this, I guess, project. Being that for this particular project, or this scenario here, we have a specific way of doing it because it's basically a quick turnover of design options with quick options for our client.

Going into Revit and linking Civil 3D information. Like this scenario before, Andrew was using an IFC method. In this case, all I'm doing is linking a Civil 3D DWG 2D file. As we link that in I've just selected origin to origin position. And the reason for this, and I'll explain that a bit later, but all we're doing now is turning on the survey and the project base point.

The survey is the triangulation, triangle icon, and the project base point is the circular one. What I'm going to do here now, after I check that, select the DWG by going to manage, coordinates. And all we're going to do is acquire coordinate positions. We select the DWG file. Once I've clicked that, don't if you noticed, but the survey point has moved now, so that's acquired the coordinates. I sometimes acquire the coordinates twice, because it brings up a warning file for me. Oh, sorry. A warning message and that kind of tells me, right, I've done it correctly.

Zooming back into the project point location. And the best way to explain this is the unclipping and clipping of a project base point is-- the analogy I like to use is like a steak and a knife. If you clip it you're basically stabbing the steak, so if you were moving a knife your project moves with it. By unclipping you're lifting it up, moving your point, and then reclipping it.

Now we've marked the same location where Andrew's put that reference point. What I'm doing now is putting a boundary line in here, so this is another method of checking because once I've drawn my boundary line I can go into an annotation and simply tag that boundary line and it's going to give me my bearings. Obviously It's the same information there. So that's a method of checking.

A couple options in that place. The reason why I tell my guys to just link in from origin to origin is it's always in the same position. I don't want to have to tell my guys, this scenario you might want to consider moving the DWG before you acquire the coordinates, because that'll just generally confused them sometimes. But this is the best method I've worked on and bringing this information in.

The other scenario-- I think the video stopped-- was, obviously, the other one is rotating your project, or the alignment now to a project north. So it's just as simple as just rotating it, which is still running. Basically you've got to manage coordinates again, sorry projects. And you can select that boundary line and it just rotates it for you. And it simply is, now you can basically have that box, like I said before, to start building up your project location.

Just in summary on that item, correct DWGs. You can always use a 2D file, you can use a 3D DWG file as well. But depending on your project size the 3D information's quite complex, and just slug-- really slows down your project environment. location between the reference data is a key point. Project base point, like I said, clipping and not clipping your project base point is quite critical. People just seem to sometimes forget that, because if you do move it after you acquire the coordinates it just throws out the whole information.

If an individual or a certain senior personnel that does that is always the best option. But pin it down, find any way to just literally lock it so that no one tampers with it. Checking twice is the other one. Boundary lines, checking the bearings is one option. The other item is using spot coordinates, or spot elevations, is another method.

What I generally like to do as well as when I set up my project is have one level and call it [? AHD ?] which is just 0, 0, 0 or something, or again how you best want to do it for your projects. But a level to have some reference to the 0, 0 location. Another method that you can look at using is the bridge extension tools. The bridge extension tools can import a LandXML file. I guess in this case we're not building a bridge, but what we can do is input this LandXML to build a deck, and that deck effectively builds our starting point, or a blank canvas to start doing any modeling and work on.

It has a lot of features but the one that we want to bring in, or most commonly used, is the road and terrain definition integration with the AutoCAD and Civil 3D, obviously, which is a good one. But like I said before the bridge definition, so the concrete slab itself or the deck that's what we want to use. So it's a good way of bringing in information to quickly develop your model.

Once we've got that information we can't wait for Civil 3D or the Civil guys to kind of give us that locked down design, because let's be honest that's never going to happen. So as our designs developed I've obviously turned off the Civil information. Nothing else has changed besides me developing our design further. What I'm doing now is linking in a more developed Civil model. In this case, I'm bringing in his full DWG 3D solid model, and this is live as well. In this case it's come through quite quickly.

And now it's just a visual checking to make sure that it's in place, and in the correct location. Again, using the boundary or bearing tool to help confirm that it's still in the correct position. The other item is using the spot elevations or coordinates to confirm that it's in the correct spot.

As our team, or in the buildings team, develop the information you can imagine in parallel you've got architectural structures again, services. So there's always this cascading effect when information needs to be updated. When you're working in a company like AECOM, or any company and in general really, you appreciate the decisions that you make has impact on the other guys, and the compressed time that they have as well.

Following on from that. Linking of 3D DWGs. Now do we still need to do that? This is kind of the method that we've done in the past, because in 2018 now we can actually link in NWC files. So if it's only purely for coordination, checking of design intent, or visuals of the design itself maybe NWC is a method. IFCs, which is the international foundation class, that's another method as well.