Enabling Technology

Matthew Baker and Eric Wittner of Esri provide an update of the core tools and geodesign workflows that can be used with ArcGIS software.

Jan 5th, 2012

Start From:
Player Color:

Right-click on these links to download and save this video.


00:01Let's go through an update of what we've been working on the last year.

00:03We had a lot to talk about last year and the year before.

00:06We've had a lot of core tools that were enabled with ArcGIS 10, and this year was all about what are the…

00:12…workflows that people want to embark on, using the software?

00:17What can we do to improve our tools and better enable people to do design?

00:21So the first thing we're going to think about is, well, what is our model for design?

00:25And we know a couple of them right now.

00:26The main one is, What's Now Called the Geodesign Framework by Carl Steinitz.

00:31We've seen this the last couple years; of course, the title has changed and I think…

00:34…that it's a good fit for this design, of course.

00:37We know that geodesign is software independent, so our challenge is, well, how does our software fit…

00:42…in with this workflow, with this framework?

00:45Thinking about what all the steps are here, how the whole model interacts with itself, and doubles back…

00:50…and goes on and so forth.

00:52Where do our tools fit in?

00:54Well, GIS does a lot of what geodesign can do.

00:57And as I said, at version 10 we can now sketch.

01:00That enabled a lot more design.

01:03We can create maps, we can analyze our data, we can gather our GIS, and all of this kind of stuff.

01:08This diagram is really the model that we followed, but we also wanted to abstract it a little bit more…

01:13…just to get people who don't know anything about geodesign to say, Oh, okay, I know what that is.

01:18And someone said to me yesterday, Well, really, geodesign is the geographic process, is that what it's called?...

01:23…right, that we know so well.

01:25Gather your data, run your analysis.

01:27Think of a question and answer that question using a GIS.

01:30And now how are we going to change what that question answered, right?

01:34How can we better solve this problem? Okay.

01:37So what we wanted to do then is take some workflows.

01:40This is our main goal this year.

01:41And we brought in some people from around the country, Canada and the United States, and we said…

01:46…what is your workflow?

01:52…and let's see how we can wrap our core tools around there and develop some better tools that will…

01:57…actually enable geodesign outside of core.

02:00So the first group we had was a group from Alberta called O2 Planning and Design.

02:03You're going to hear more from Doug about the work that they're doing, but let's break down what we did…

02:08…with them and how we took their workflows, wrapped them into core, or at least figured out how they worked…

02:13…with core, and then made some new tools from a couple other workflows, including theirs.

02:18So Christian and Bill came down, and we had at it for a whole week in our conference room.

02:22And what you're seeing them do here is, as Jack discussed, let's take all of our workflows…

02:26…let's diagram them out, let's see if they make sense, before we even touch a computer.

02:31We took a marker board, took some paper, thought about everything we wanted to think about, and made some maps.

02:37Now these aren't maps necessarily to communicate a lot of things to people; they're maps to say…

02:41…here's where something is going on.

02:43Here's an area I want to avoid, here's an area I want to favor.

02:46And this is the whole part of, you know, the beginning of the geodesign framework.

02:49Gather your data, analyze your data, start to answer some questions before you even get to the design.

02:55So what we ended up with is taking our good old friend ModelBuilder.

02:59Without ModelBuilder, where would we be in a GIS?

03:02Very simple thing, take some data, take a tool - a tool has a parameter - that is, it needs to know something about…

03:08…you know, the data and how the tool is going to run, and it creates a piece of output data.

03:12You can expand a model as much as you want.

03:14We made a lot of models to answer a lot of different questions about what was going on in their study area…

03:19…north of Calgary.

03:21And, of course, don't worry about what the individual datasets are here, but this is looking at habitat suitability…

03:26…for a few different species that are critical to the area north of Calgary and their study area.

03:32Some of the maps that we ended up with was this one, looking at the vulnerability.

03:36User areas that we want to avoid.

03:38So this is a combination of a lot of different datasets.

03:40And you can see that it follows some natural features.

03:44There's some other sort of blockier things in there; it just depends on what goes into that model.

03:47And here's the output.

03:48We wanted to get everything on a scale of one to nine.

03:51And what that means is all the data that we put in there - roads, habitat, wetlands, whatever it is…

03:57…let's break all this down and get it into a common language on a scale of one to nine.

04:02So if you're familiar with how ModelBuilder works and suitability analysis, this is a common thing.

04:07This is part of the GIS process, right?

04:09So this is one of many maps that we created.

04:12We also ended up creating some more constraints for development.

04:15These are some more areas that we wanted to avoid - terrain, habitat, ecological infrastructure, agricultural areas.

04:22What does this now become?

04:23Well, this becomes maps that we put into our document that we can see.

04:27When we design, we avoid them.

04:29We are making an intelligent design decision, because we know the answers to these questions up front.

04:35Because we're using ModelBuilder, it's a very dynamic little piece of software, again, that allows you to take your…

04:40…data, your tools, your outputs, keep your workflows going, but then go back and make any changes if you want to.

04:46So it's a living kind of diagram, right?

04:49And it creates all these datasets for us and again, we use them for our intelligent design.

04:54So what do we then do?

04:55Well, you've seen in years past, you know, we have the whole sketching environment here, right?

05:00The sketching environment allows us to create data.

05:01We can see where the areas are we want to avoid.

05:07The sketching palette on the right-hand side allows us to choose the type of feature, draw it on the map.

05:12Remember that we're not drawing pictures, we're drawing data.

05:15And so we're going talk about how to evaluate that design, not so much as, well, I can see that it's outside of the…

05:20….constraints, but what does it mean?

05:22How big is it?

05:23How many of this does it contain?

05:25What does it overlay with? Okay.

05:28What we're also doing is we're taking our Wacom tablet, and we're using it to enable sort of a…

05:36…free hand, free-flowing sketch of our plan.

05:39So this is nice for, you know, here's the broad plan, here's what I think it should be.

05:43You know, like I said, we've seen this in years past.

05:45But it's a very friendly, user friendly way to create a nice-looking plan and of course because it's data, it's accurate.

05:52Right? It's got information about it that we can use.

05:55Now I want to say this. I'm not going to have this on stage this year.

05:58We are going to have a sketching ideas lab tomorrow at 1:30 back in Room D.

06:03So we're going to open up the technology, we're going to open up this Wacom…

06:06…all the sketching palettes, and the new tools that I'm going to show you in a moment that we've been developing.

06:11So let's keep going here.

06:15What the design has ended up with, some mock designs that we came up with, are actually very interesting for me…

06:20…as someone who's interested in, you know, new urbanism and urban planning and this type of thing.

06:25In years past, we've said, Okay, well if we're going to just sketch out a land-use plan…

06:28…we're going to do it using land-use codes.

06:31But then we thought, well, eventually we're going to move this into our CityEngine software…

06:35…which I'll discuss in a moment.

06:37I can have commercial areas out in the boonies, I can have commercial areas in the urban environment…

06:43…but they're both commercial if I use only land use, right?

06:46So we decided to create another layer called Intensity.

06:49And this was the first layer that we're going to sketch out in our plan.

06:52And it describes the urban down to the natural, the industrial to the institutional.

06:58Basically, how many people are going to live in the different parts of my design?

07:01And of course symbolized by the different shades of green; we have civic and industrial in purple and blue.

07:06The intensity plan's our first layer.

07:09Next comes the land-use plan, which kind of mimics the intensity plan, but now we start to map out where…

07:14…the different land use is going to go.

07:16And again, both of these plans are created using our sketching environment; grab a symbol and draw it on the map.

07:21But now we know where we're going to plan all these things.

07:23And you can see that the natural features that we've drawn again enabled by that tablet…

07:28…which you can do with your mouse, but a little bit easier to do with a tablet.

07:32We're starting to see now a bit of a town develop, right.

07:35We've got some streets in here, we've got some urban/commercial, urban/mixed use and getting away…

07:41…into the residential again into the open space as you get away from there and some agriculture.

07:49Another important part of doing geodesign with ArcGIS is evaluating your design.

07:54So in years past I’ve shown a small window that, you know, shows the chart and updates as you edit…

07:58…evaluates some kind of indicator and plots that out in a simple chart.

08:02We want to be able to extend that, and the one way we're doing that now is with an…

08:05…Excel-based dashboard manager.

08:08We're using Excel because we have tools that enable the communication between ArcGIS data and Excel.

08:16So think about all the things that ArcGIS can do, think about all the visualization, expression…

08:21…evaluations, and all these things that Excel can do.

08:24If the two can talk, that means that I can easily go in and create little buttons, little sliders, little charts…

08:29…and all these things, and actually also read parameters from ModelBuilder.

08:34Click a button and rerun a model in ModelBuilder, and Excel reads the data again.

08:40This is very handy, because I can, if I know Excel, I can extend this, right.

08:45I can add my buttons, I can do whatever I want, and it's a very broad platform for creating custom dashboards.

08:51And so this is what we're developing as well, again, using the O2 workflow we figured, well, here's…

08:56…how our core software works, here's how our dashboards work, and here's where we want to go…

09:00…with this type of evaluation.

09:04So once we figured out that we had a good design, we'll say we had a good design, you guys had a good design…

09:09…we wanted to bring that into CityEngine.

09:11Now CityEngine is this software that was acquired this year; I don't know if you've heard about it yet…

09:15…but as the name describes, it is an engine for building cities.

09:20Now we kind of sat around and scratched our heads and said, Okay, well, we're going to do our intensity plan…

09:25…our land-use plan, we're going to throw roads on our plan.

09:29Now that's really all we want to design in ArcMap before we pass it off to CityEngine.

09:33CityEngine has a lot of great things that allow you to generate what's in between those roads…

09:39…and read your land-use plan and your intensity plan.

09:42So we came up with a workflow that said, We'll take these GIS datasets that we created in ArcGIS…

09:48…put them into CityEngine, generate the roads in between the main arterials.

09:52Generate the parcels in between those roads, and then put houses or whatever buildings go on those parcels.

09:59But the neat thing is, the building heights, the building sizes, and the building types, as you can see by the color…

10:04…are all derived from GIS data that we created in ArcMap.

10:08All right, so the orange color, the yellow color, the light yellow, those are the residential, the mixed use, and so on…

10:13…that we designated in ArcGIS.

10:16And now we're talking in CityEngine, it has literally generated the city based on this rough plan that we drew.

10:23Now the goal is to more seamlessly integrate these, right?

10:27Well, again, open these tools up in the sketching lab and see how they work, but we want to be able to say…

10:33…look at this right here and say, mmmm, I'd like it to be something else.

10:36Go back into ArcMap, make that change, see how it updates in CityEngine, see how the buildings change…

10:41…see how the sizes change, the heights, and all these types of things.

10:45And really how CityEngine works is, it allows you to specify rules.

10:49And now I'm going to say, Well, based on this land use and this intensity, I want these houses to look like this.

10:55I want them to be this tall, this color, have these shadows, and all this kind of stuff, right?

11:00You can see, you can really get down into a neat little picture of what your community might look like.

11:06And so the rule that we've created here says, It's going to be this dense, it's going to have these buildings…

11:11…and this is how it's going to look.

11:12So CityEngine is very exciting, and to see the plan that we created in ArcMap come to life in CityEngine…

11:16…was very exciting for us.

11:19So let's then think about what we had to do after we had all of these workflows.

11:23So we had O2 come in, we had a group from Toronto come in from the University of Toronto…

11:27…and we had a group come in from North Carolina, and just several other communications that we had…

11:32…throughout the year.

11:33That again, just gave us an idea of what these workflows are.

11:37Let's see if we can identify some patterns, and of course software development here at Esri is user driven.

11:43So whatever the users want, this is what we want to put into the software.

11:47So the big thing for us this year, and I don't know if you can really see it here, is a new window for geodesign.

11:51And there's a couple different things that this window does that enables not all the modeling and all the stuff…

11:56…that you are going to do before you get to your design, but how do you enable the sketching interface now?

12:02How do you enable a smooth transition into a design?

12:05And like Jack was saying, there's a lot of content out there on ArcGIS Online, and we're starting to use this…

12:10…as a central place to go out and get things like templates.

12:14And we thought, okay, well, let's take that template that we created with the O2 folks…

12:19…and let's build a window that allowed us to go out to ArcGIS Online and pull it down into my…

12:25…ArcMap document and start doing some sketching.

12:27So this window right off the bat will allow me to sign in, or not, to ArcGIS.com, put in some keywords…

12:33…find some kind of template that I have out there and pull it in.

12:37And then the tools behind the scenes will just wrap that, or sorry, unpack all that information…

12:41…open up your sketching interface, and allow you to just go ahead and sketch.

12:46With the core tools we have the feature templates. With this geodesign window...

12:51…it's more of a design-oriented thing.

12:53Feature templates allow you to create data.

12:55We want to do design, right?

12:58So you can see the feature templates.

12:59There's our land-use palette that we had; there's the intensity palette that's in there, and all of these types of things.

13:04Our feature construction tools are in there.

13:06How do we want to draw?

13:07Do we want to draw a line, a rectangle, use free hand, or anything like that?

13:11So that's in there as well.

13:12Again, we're just driving bits of core.

13:14'Cause as I said, ArcGIS can do almost everything that we want to do in geodesign, but the tools are everywhere.

13:21Our challenge is to create a nice interface for doing design in ArcGIS, right?

13:27Not just, well, not any workflow, but so many workflows that it can handle.

13:34What kind of tools do we put in there?

13:36Well, there's tools that designers want to use when they're doing sketching.

13:41Remember that we're using GIS data, not just graphics and simple little rasters.

13:46So, can we build these tools around GIS data?

13:49And sure enough, our Charlie McCloud, our very talented developer, has been working hard to get this window going.

13:55And again we'll look at it in the sketching ideas lab.

13:58He created a couple of tools that are so cool for doing design.

14:02So we can sketch out all these things and then we can select, say, I want to know what thing is on the map…

14:07…tell me what it is.

14:08So I can click on it and it shows me what it is in the template.

14:10I can click on a feature in the template, and then it will show me all those features on the map, which is great.

14:16I could also now, well, I guess we've taken what was in ArcSketch, if you remember that, the ability to…

14:22…select a feature and engage the grab handles and rotate it and skew it and move it around and all these types of things.

14:28A very easy interface for manipulating the features you've created.

14:31And again, this is all enabled by this geodesign window.

14:35And another cool thing that he did is created this paint bucket tool.

14:38So if I have a sketch or a plan that I've designed, I can simply select a new type of feature and just drop…

14:43…the paint bucket on those features and they all change to the new type.

14:47So very simple things, but very effective, I think, for doing design.

14:50And again, let's get some feedback from you all here while you're here.

14:53Come to the ideas lab, or stop me or Eric or anybody and we'll get this going.

14:59Okay, this is something that's very exciting for me.

15:01So you'll read in Bill's white paper that the scenario management is a concept that doesn't really exist in a GIS.

15:07We spent the whole year trying to think about, well, what is it? What is it?

15:10And then I thought, well, it's easy, it's just a piece of text that describes your scenario.

15:15And that's what we did.

15:16So we built another field in the attribute table, and we thought, okay, we're going to put the name…

15:21…of the scenario in that field.

15:23How do we then interact with it?

15:24This is where the…I'll call it magic 'cause it is magic, but anyway…

15:29…I can create a new scenario, and really I'm just creating a name, a piece of text, that when I draw…

15:33…that feature, it gets that scenario.

15:36And it also says to the other features in the map, hide those features and only show me the features…

15:40…that display that piece of text, right?

15:42So if you know GIS, it's a definition query on the map, and it's a feature template default attribute…

15:47…so I can sketch with that scenario.

15:49So what you can see in that red box there is, there's a toggle, so to speak.

15:55Where I can toggle between my different scenarios.

15:58And so the ability to wrap your whole plan, multiple scenarios, into one file is extremely powerful for us now…

16:05…because we want to store all of our scenarios into one file, maybe send you just my scenario…

16:10…send you the whole design, whatever it might be.

16:15Of course, the back end GIS tools will allow you to store the date of a feature and anyone's name.

16:20So I have scenario name, I have the date, I have the person's name, and we thought, well, with that date…

16:25…wouldn't it be nice to take that design somewhere else and put a time slider on it, so I can see…

16:31…how that design evolves maybe over the six months or eight months that it was done.

16:36So that's all possible with this window.

16:38So this is very exciting for us. Okay.

16:43Jack mentioned a little bit about the web, and I won't touch on this too, too much, but we can certainly open it up later.

16:49And I was just going to show a little picture here of Explorer Online.

16:53Now what I did is, I took what we did in ArcMap and simply threw it up onto an ArcGIS Online hosted service.

16:59The ArcGIS Online hosted services, I don't know about you guys, but I don't, well, I don't like installing…

17:05…ArcGIS Server, and I don't want you guys to have to do it necessarily.

17:08But, hosted services, you don't have to worry about having a web server, installing Server…

17:13…having an SDE and all the back end tools that enable web editing.

17:17Taking your map and throwing it right up onto the web and clicking a check box that says Enable Web Editing.

17:23Done. Now you can bring it into Explorer Online or any other map viewer, like Flex, JavaScript, Android…

17:29…iPad, whatever it is, and enables, this enables collaborative web editing.

17:34Because everybody's got, you know, access to this plan, it's in a database on the server.

17:39I can open it up, make a change.

17:41It goes right to the database that somebody else can see, they can make a change, they can evaluate it…

17:46...they can bring that in and use their dashboard on it if they want to evaluate it, or somebody else can evaluate it…

17:51…on their parameters, as well, enabled, of course, by ArcGIS hosted services.

17:57So what's next for us?

17:58So again, feedback is the biggest thing.

18:01We don't really have anywhere to go unless you all say, This is where we want to go, and we're listening, right?

18:06So we gathered some feedback this year, and this is where we're at with the tools, and we want to keep going.

18:11We want to have as many of you who are interested in the actual tools come to this ideas lab tomorrow, 1:30, Room D.

18:17We're going to have Wacom there, we're going to have myself and Charlie and a few other people from Esri…

18:21…and as many voices as can make it to just open it up and let's hack at it and see what we can do.

18:27There's a group on there, on ArcGIS.com, if you're interested.

18:30I'm starting to throw up little add-ins, little feature templates, and any other little tidbits that I can…

18:35…get into your hands to enable geodesign.

18:38So there's a group on ArcGIS.com called Geodesign with ArcGIS.

18:42And I want to thank you for my time and send me an e-mail if you want, and I'll throw it over to Eric.


18:56So hello, everyone. I'm Eric Wittner.

19:00My emphasis here at the company for the past six years has been 3D.

19:05And so what I want to show you is actually some of our software, some of our technology as it can help drive…

19:10…design in the urban environment.

19:13As Shannon mentioned, we have more and more people in this world, and we have less and less space…

19:18…that we want to put them in, as everybody seems to be moving to the urban environment.

19:23So what I want to first introduce you to is just kind of the technology that we use to represent the urban environment.

19:30How we build our representative model, and there's been a few advances in the last year that I want to highlight.

19:37Some of you've probably seen portions of our technology that we used to have.

19:42Of course in 3D or at the basic component is, there's imagery, high-resolution imagery, to tell you where you're at…

19:49…it's draped on the surface of the ground, which gives you your elevation, and then it's populated with…

19:55…various types of building information, and the humblest piece of building information is of course the footprint.

20:03And a footprint can be made more interesting because you can make it 3D by extruding it by attributes.

20:08But one of the kind of critical types of data that's emerging in importance that's becoming actually almost more…

20:14…kind of ubiquitous is lidar.

20:16And a lot of you probably have different types of lidar information.

20:21And if you have access to first return, you can actually derive the heights of buildings from your lidar….

20:26…so you can actually use that as a way to kind of very rapidly create an urban environment.

20:32And in the last year we've really greatly improved our support for lidar information.

20:37So what we're looking at here is actually full-color lidar, 150 million points for the city of New York.

20:46And we can actually zoom in, and we can use that to guide the construction of geometries, the validation…

20:51…of geometries, or just to kind of investigate what the world looks like in 3D.

20:58So here we're looking at a zoom-in of that section.

21:01We can densify it and again, since this has color associated with it, we can take a look and see what it looks like.

21:09We can also do measurements off of it.

21:13But really what I've talked about there is mostly just visualization. It's also, we can do analysis against it.

21:18So, here we've got a collection of lidar.

21:21We can run a line across it and get a little profile of whatever we're looking at.

21:27We can reach into that lidar information and manipulate it, so we have a collection of trees.

21:32We're just going to grab those and turn those into vegetation, 'cause…

21:37…they look like tree canopies.

21:38Right, and that's reflected on the ground.

21:41And I think where we want to go in the future is actually, at least it seems to be what you guys are asking for…

21:48…is feature extraction.

21:50So if we look at it in a 3D viewer we can see we've got a collection of points that clearly, they're…

21:55…rooftops of buildings, wouldn't it be nice if I could just assign those as buildings and extract those.

22:01We have kind of a dynamic classification tools now where you can go in and define a range of elevations…

22:08…grab a point, pull everything out that participates in that, and assign it a sim attribute.

22:14So here we're just turning that on the roofs.

22:15But the next step for us is to actually to consolidate that into a geometry and give you a shape out of it.

22:21So that's kind of a new, exciting kind of support for a different format to populate your urban environment…

22:27…that we didn't have up till now.

22:32So another thing that I, you know, I showed last year was we approved our support for kind of…

22:34…interaction with third-party models.

22:40So if we had a simple extruded building we could throw that out and bring it into [unintelligible] or SketchUp.

22:46But a lot of people who are working in the design profession are working with the kind of much more complex data…

22:51…that they're getting from BIM information, building information modeling.

22:57So what I actually have here is, this is a IFC file that was generated from Revit, and we've imported it…

23:05…using our data interoperability tools that our partners, Safe Software, helps us put together…

23:11…and brought it into GIS and georeferenced it. I just dropped it into the city of Philly.

23:16But so you can actually start bringing in this kind of detailed, detailed building information…

23:21…if I turn off the walls and the ceiling.

23:25We can reach in; we can actually see the room types inside the buildings - the doors, the walls, the windows, et cetera.

23:31So this is another new data source that you can use to start populating your urban environment.

23:39Matt mentioned CityEngine, and he showed kind of a model of buildings generated based on…

23:45…that was meant to articulate the land-use type and kind of proposed density.

23:52That tool is capable of making actually quite complex geometries based on that kind of same rule set.

23:58This is a building that we generated in CityEngine and brought in, and I'll talk a little bit more about that later.

24:04But really what I'm trying to highlight here is that we now have the ability to take all these pieces of data…

24:12…from SketchUp, from IFC, from extruded buildings, from lidar, from models that…

24:22…you may be able to generate and bring them into one centralizing place, right?

24:25And then share that back out with other software packages, as well.

24:30So kind of, we always get pressured to take kind of a coordination role, and I think this year we've…

24:37…kind of made it over the hump in being able to effective to provide that to you.

24:42And it's not just visualization information, right?

24:44We can query these features and get their attributes.

24:48We could see if they have floor plans or who's the tenant in a different building…

24:54…or when was that building built?

24:56So it's actually an intelligent information model.

24:58It can dynamically label itself, so that you know what you're looking at in 3D.

25:05And since it's in the GIS, right, it's right there parallel to your existing 2D GIS data.

25:13So that all can play with kind of your 3D urban design.

25:17So in this case what I've done is, I've just taken the buildings and pulled the zoning codes onto the buildings.

25:24But this could be condition of building, it could be a variety of other attributes, like response times…

25:30…I'll show you an example of that in a minute.

25:34And it's not just, we've talked kind of at the building scale, but it actually gets down to much more detailed information.

25:43So if I zoom in here, we actually have pretty detailed information for this section of Philadelphia.

25:51We have street trees information with heights, canopies, ages.

25:56That can be driven actually from two perspectives.

25:59One is you can have your existing street tree coverage, and the second one is you can have your proposed.

26:04And you can track the cost of the tree or the size of the canopy and the water usage over time.

26:10And use that when you're doing your design to drive the cost of the year's design.

26:13How quickly are you going to get to the benefits of planting those trees to take sun off your sidewalks…

26:19…or reduce heat islands effect.

26:21What is the total water cost going to be over time?

26:23What is the total maintenance cost of that vegetation over time?

26:26So again, it's developing these kind of robust data models behind it, such as our city data model or our local government…

26:31…data model, our campus and facility data model.

26:34They're going to inform these kind of designs and build in these metrics, so you can actually do…

26:38…these measurements to help you do good design.

26:43One of the other things that we see here is street furniture.

26:46Again, that might have a whole series of attributes.

26:48It might be existing condition.

26:50So here I've color-coded the streetlamps based on what the condition is, maybe how much graffiti it has.

26:57But from a kind of planning perspective, you're probably more interested in the specs of what it is.

27:01How much light does it cast and where does that light go?

27:05So you could actually, based on the type of light, generate different types of spheres or cones…

27:10…showing where that light's going, and use that as you're doing a new design to plant streetlamps…

27:16…in different locations and see, am I getting effective coverage?

27:19Is this area going to be secure when I actually put in these streetlamps?

27:26And we've been talking about the outside shells of buildings, we've talked about the things that lie on the surface…

27:31…but actually we're going beyond that level of detail now.

27:35So across this square I have a little building called the Franklin Institute.

27:39If I get close to it, you can see the exterior turns off, and I'm actually looking at the interior space of this building.

27:45And you could attach all kinds of information, all kinds of asset information, tenant information, et cetera, to this.

27:52And you could actually do emergency routing. This is a 3D network that we could do emergency preparedness routing.

27:57We could test how far can a hose reach into that building.

28:01Can we actually get a 200-foot hose to the back auditorium to put out a fire, in case there's a fire.

28:09So these are all things that kind of are, in the physical world, that are readily seeable.

28:14Let's say we actually had an accident, or an emergency on the street corner.

28:18What I want to know and understand is, where are the people who can respond to that emergency in my city?

28:24And it's not necessarily easy to recognize where your emergency responders are, just based on the type of building.

28:32But we have this kind of ability to do dynamic labeling.

28:35So here we have fire stations labeled in 3D.

28:37If I zoom in on one of these fire stations, I can see it's got an engine company and a ladder company.

28:44And what I want to understand is, of these fire stations that I have, which ones can respond to…

28:50…my emergency here at the Franklin Institute within four to five minutes?

28:56So they're in a realistic response time for an emergency.

28:59So what I've done is, I've created a model, and that model looks at all the fire stations…

29:05…finds the fire stations in a five-minute drive time, then it does the drive analysis to show what…

29:13…the response times are in color, and then it actually takes those and applies those to the buildings so…

29:19…you can understand the response time in the surrounding buildings.

29:24So I'm not going to actually run that.

29:25One thing I should point out is actually the drive-time analysis, in this case, is not something that I have on my computer.

29:32That's hosted in the cloud.

29:33I'm pulling a tool that's living in the cloud, making it a component of my model with the tools that live…

29:39…on my computer and using them together to answer a question.

29:44And I think this is really what we as a community need to start working on, is building these tools…

29:49…these geoprocessing models that kind of enable analysis, and publishing them as services or publishing…

29:57…them as geoprocessing packages, or publishing them as templates out onto the web that we can then share…

30:03…so that we learn from each other and we improve our ability to make decisions in a kind of an easy way.

30:09So here's that drive-time analysis.

30:11Let me zoom out and you can see the two fire stations, and this right here is where our emergency was.

30:18So we can see we just barely eked into the three-minute response time.

30:23And if we wanted to, we could actually use that to drive decision making.

30:26So one of the new things, I know location-allocation is one of Jack's favorite tools.

30:31There's a new capability, and it's coming in the next release, that actually allows you to do location-allocation by capacity.

30:39So you can assign each fire station the total number of people that they can serve.

30:43And then they create their drive times and their distances of who they can serve, not just based on how far…

30:47…they can get, but when they meet the maximum number of people that that can group can effectively serve.

30:53So as you're doing design you could actually compare that, and say if I increase density massively in this area…

31:00…based on the existing configuration I have of fire stations, do I have enough emergency responders…

31:05…to support those people?

31:06And if the answer is no, then you have to build a new fire station.

31:10So here's that same data, except in this case, we brought it onto the buildings.

31:17So we're pulling those attributes for emergency responders and displaying them in 3D on the buildings.

31:25So we can actually use it to derive 3D visualization, thematic visualization.

31:33So the next thing I'd like to show you is actually another piece of information that you can't readily see…

31:41…when you're above the ground.

31:43So what we have here is, we have the subway system for the city of Philadelphia and all we've done is…

31:49…draped the line on the ground, and we've kind of highlighted where the stations are located.

31:56And so underground information is definitely one of the things that a lot of people are interested in planning…

32:01…especially for conflict detection when they're dealing with utilities or transportation and they're trying to…

32:06…propose designs and interacting with those buildings.

32:10In this case Penobscot Bay Media actually gave us an interesting dataset.

32:17I'm going to actually zoom down into one of these buildings.

32:22And we actually have some more detail for one of these subway stations.

32:26So let me turn the imagery in the buildings off real fast so they don't interfere.

32:38So, if I tilt underground what you're going to see is we have the subway station in 3D, or the components of it…

32:46…that were collected underground.

32:48And they're color coded.

32:51The blue areas are transitions to the surface.

32:53The green areas are boarding areas, the gray areas are walkways, the yellow are ticketing booths.

32:59So what we actually have is, we have thematic land use underground in 3D.

33:06And I think this is definitely going to be one of our pushes in the coming year, is that we're hearing a lot of people…

33:11…who are interested in doing their planning work no longer at the 2D but in the 3D.

33:16And so they want to do on the first floor commercial, second floor commercial, third through sixth floor, offices.

33:23Above that, residential.

33:24And then be able to take takeoffs and say, How many people are coming into this building at 8:00 a.m…

33:28…how many people going out?

33:29What's the water usage? What's the electric usage? What's the carbon, total carbon footprint of that building?

33:35Driven by that kind of 3D data model of land use or space use inside that building.

33:41And that can be at this kind of floor-level scale, or it might be at the suite-level scale.

33:46Or in this case, it's actually as detailed as the room scale.

33:50So all this information actually was collected using a robot, and that little robot runs around and takes pictures as it goes.

33:59One of the cool byproducts of that is that you get this kind of great piece of data for virtual reconnaissance…

34:07…and planning if you're an emergency responder.

34:10It's a spherical image that you can look in 3D and actually see what that space looks like, wherever that unit went.

34:18So again, the potential of a planning tool for emergency responders collected as a byproduct of that kind of lidar collection.

34:28So what I'd like to do is actually walk you through a quick little analysis routine, and a little planning routine…

34:36…for a new set of proposed buildings.

34:39So what we have here is land use; it's just represented in 3D as extruded polygons.

34:45And the city decides that this kind of section of area over here is underutilized.

34:49What they've done is gone through and identified parcels that are either abandoned…

34:54…which are the dark yellow, or suitable for redevelopment, which are the beige.

35:00And they want to understand, based on where they have the available parcels, based on their land use and based on…

35:06…proximity to the stations for the subway system, where they could potentially redo redevelopment in this area.

35:14So again, what I've done is, I have a second model.

35:22And this model does, basically, a weighted overlay.

35:25A suitability analysis between those things and it calls, again, an external service that's living on the…

35:30…Amazon cloud somewhere to do walkability analysis and creates for me a new redevelopment suitability model.

35:40So we can see this is the suitability for kind of commercial development in the area.

35:45We have high suitability on this bottom right side, because we've got good colocation…

35:51…with these Metrolink stations.

35:53But we have kind of low ability on this side, where we actually want to do our redevelopment.

35:58So we make a decision that what we want to do is actually add in some new light rail systems…

36:03…some connection to downtown. Maybe instead of a surface rail system, we're going to do a subway rail…

36:09…system or an elevated rail system.

36:13So what we're going to do is actually decide where we want to put that in order to service our area.

36:20We could use the sketching tools that Matt showed to quickly sketch in a route.

36:24We want to understand if this is a good idea to build this route.

36:27So what I'm going to do is quickly tilt underground so we're looking up from below.

36:35And I can take that proposed tunnel that we built and turn it actually into a piece of geometry.

36:41So this is actually a subway tunnel; it's the correct size.

36:45And it turns out for this city we actually have soil information - boreholes.

36:50So we know what type of soil is underground.

36:55We can figure out which pieces of that soil intersect with the tunnel and then actually use the tunnel as a way…

37:01…to kind of clip the soil so we can look at and see what is this tunnel made of?

37:06What is it made of at different depths?

37:08And again, since it's GIS information, we can ask it a question, right?

37:12So we can look at the volume and can say, Oh look, you know, it's mostly clay…

37:16…that should be less problematic to remove.

37:19So yeah, we definitely want to add in this new transportation corridor.

37:26So once we have that corridor, the next question is, Where do we want to put our station?

37:30And since this is kind of the area we're targeting, we're going to put in a train station right there.

37:37And we want to rerun our analysis to see how that's impacted our suitability.

37:42So I have, again, just take the same model; rerun it again, except use the new stations.

37:47And you can see, all of a sudden we have much higher redevelopment potential for this area.

37:53So we want to go into this area and actually start sketching in some proposed buildings in the area.

38:01And we could do that in GIS, right?

38:03We could draw polygons; we could extrude them, make really simple massing models, get square footage.

38:09We could have a model behind it that gives us electrical usage, water demand, potential impacts on traffic.

38:16And, I think the most impressive thing here is that these kind of simple massing models that are just…

38:20…absolutely so gorgeous and compelling when you show them to people.

38:25So this is why I am really, really excited that we acquired Procedural this year, because what you can do is…

38:31…you can take your land-use plan and you can take your proposed design and you can push it across…

38:36…into CityEngine and make it more attractive.

38:39And we have some new templates now that allow you to actually take this…our 3D city model…

38:44…and our campus model and push them across into CityEngine.

38:48So what I want to show you now is CityEngine.

38:51This is the same city; I've brought in the contextual buildings.

38:54And what I've done is, based on a rule set, I've generated some buildings for our proposed area.

39:00This is the kind of area I was focusing on there.

39:02We've got some skyscrapers and then we've got some kind of more townhouse-scale buildings and this location here.

39:11And these are rule-driven buildings.

39:13So let me give you an example.

39:15I'm going to select this building here in the foreground, and I'm going to make it…

39:21Let's make it drastically taller.

39:27All I did was change an attribute in the GIS and boom! We get a new building.

39:32I don't like that roof top; I want it to be a pyramid, and I want the pyramid to go five meters higher…

39:39…than the rest of the building.

39:41So these are all things that you can design in your land-use plans and your intensity plans that can then…

39:47…drive the generation of this 3D geometry.

39:50This is a pretty simple building.

39:55That's a bit of a tall steeple there, wasn't it…

39:59And this was being driven from footprints.

40:01So in this case you're authoring footprints, you're drawing the specific shape of the building.

40:06One of the more interesting capabilities of CityEngine is that actually, what Matt was alluding to…

40:12…was that you only need streets, and CityEngine can fill in the gaps.

40:17So CityEngine can say Okay, you have these streets, you're on a low-intensity land use.

40:23You should have lot sizes of this range, and then automatically populate the interior space with parcels.

40:31Then on those parcels, it can say, Based on intensity and land use, I'm going to derive a building…

40:36…that matches your requirements.

40:37And here's, the thing that's amazing about this, is that you can drive a whole series of type of…

40:42…zoning and planning codes into that.

40:44Setbacks from streets, setbacks from adjacent buildings, maximum height, maximum height based…

40:49…on skyline analysis, to create these kind of very dynamic buildings.

40:54So, let me go over here and we'll look at this pretty massive skyscraper here.

41:01So I can just grab this feature and again, just like the last one, I can very, very quickly change the total number of floors.

41:09So now it's a very, very short building.

41:12I can change the height of the ground floor, I change the total height of the floors in between.

41:17Again, that could all driven by attributes in the GIS database.

41:23If I make that, let's see, forty floors again, so… pop that back out.

41:28There's a style selection menu.

41:30So, CityEngine comes with these different kind of concepts of buildings, and you can kind of quickly change…

41:35…back and forth between them to change the style of the building you're building.

41:39So now we've moved to a L-shaped building that is facing forward.

41:43I could turn it back into an L-shaped building that's facing the front, right, so the L's wrapping around to the front.

41:49It's actually populating the green space on the ground based on a ratio that you set in your GIS database…

41:55…on how much green area that needs to be on the ground compared to the building.

41:59I can change the setback, so if I make the setback from the road bigger, the building shrinks.

42:04If I make the setback from an adjacent building bigger, the building shrinks.

42:09Now one thing that probably isn't apparent is that as I mentioned that the lots here are actually driven by the streets.

42:17So if I go in and I select the street, and I actually move that street, I need to actually get the street…

42:32…both street segments…there's the street.

42:47So if I grab that segment and move it around, my buildings actually change.

42:52I need to grab more segments.

43:05Oh, I'm rotating, actually.

43:09Okay. Yes?


43:17Oh, I'm on a rotate, aren't I?

43:20There. And now the whole parcel resized, all the buildings changed, right, so again, you can kind of…

43:26…dynamically change the landscape, change the roads, and the parcels are being generated on the fly.

43:33And the great thing is, is now that they are no longer a business partner but they're acquired…

43:37…they have strong integration with our technology.

43:40So I can take the building that I generated in CityEngine, bring it back into GIS, and represent it here.

43:50So here we see that sketched building with the other proposed buildings that we had living in our GIS.

43:56And now we can actually use that in analysis.

44:00So what I'm going to do, is do some line-of-sight analysis.

44:06So what the city has done is identified several views in the city of Philadelphia - this park here, the Franklin Institute…

44:17…there's a park over here, the archdiocese - these are things that people want to see, they're attractive views for developers.

44:28So what we want to do is understand how many of these attractive views can different segments of this building see.

44:35So maybe your developer wants to be able to kind of estimate cost or value based on that.

44:41So we've developed a little routine that actually takes that building and splits it into pieces and then creates…

44:48…a series of sample points along the surface of the building and does line-of-sight analysis between those two sections.

44:57And you actually get a visual representation of how good the view is from those various locations.

45:06So we can see here a surface - if I turn off the panel points it's easier to see - where we can see the…

45:13…quality of the view for the building in 3D.

45:16So you can see the people on these floors have low-quality views.

45:19Red is low quality, green is high quality.

45:22You can see that these windows right here have less views because they're blocked by the building itself.

45:28The people on this balcony actually have less views above this balcony because the balcony itself cuts them off.

45:35Right. So a way of doing kind of cumulative view analysis using some of our tools.

45:42Another real critical component of view analysis is skyline analysis.

45:50So, what we have here is a view from the steps of the museum here in Philly, and we're looking back at our…

46:01…developed area and we've got our proposed building in the foreground.

46:05And what we want to do is generate a skyline so we understand what the critical components of that…

46:10…skyline are, so we don't violate them.

46:13And this is actually kind of an intensely 3D analysis.

46:17So here if we zoom up we can see that this skyline actually - I apologize for moving quickly, nobody get dizzy…

46:25…we can see that the skyline actually takes a component of this building, and then reaches off into…

46:30…the distance to get a tiny piece of this building.

46:34Then it's composed of this building that reaches foreground.

46:36So it's actually a very complex 3D shape.

46:39Much more complex than people think.

46:40People usually think of it as a flat thing, that you draw a line on, right, but it's actually much more intense.

46:46And we can take that and we can turn that into an analytical surface, and we can use that surface…

46:52…to compare against proposed designs.

46:54So here, if we kind of tilt underneath, we can see that our proposed building just barely squeezes and in…

47:00…and doesn't violate the building behind it, so we're not going to negatively impact our skyline for the city.

47:08And then kind of the last thing I wanted to show you is just a brief kind of emergency response idea…

47:16…is that there's a huge volume of GIS tools that are out there, that you could combine together…

47:21…to do a variety of remodeling.

47:23There's a whole set on hydrology.

47:25So what we've done here is actually modeled how much the area would be flooded if there was a…

47:34…certain amount of flood flow coming down this river.

47:36Philly has this wonderful, kind of below-grade road that looks like it would be a nice conduit for water to go in.

47:43And then we can assess which buildings are going to be in the flood zone, which ones we might have to do…

47:47…emergency response for, again, use that to drive design and whether we're going to put basements in those areas.

47:54So, just an example of some of the 3D technology that we’ve been working on in the last year to help kind of enable design work.

Copyright 2016 Esri
Auto Scroll (on)Enable or disable the automatic scrolling of the transcript text when the video is playing. You can save this option if you login


No comments. Be the first to write one below.

Comment on this Video