Interviews Planning in Three Dimensions
Written by Matt Ball Sunday, 23 August 2009 18:36
While urban planners may underutilize GIS, there’s growing momentum to harness the toolset for richer visualization and analysis. Niels la Cour, physical planner with the University of Massachusetts Amherst, is doing some innovative work with 3D modeling for campus planning. V1 editor Matt Ball spoke with la Cour about the evolution of GIS for planning, the use of 3D models, and the potential advancements of GeoDesign.
V1: You’ve been involved in using GIS for planning for quite some time, has planning always been your focus with this toolset?
la Cour: I’ve been a planner for more than twenty years. Because I couldn’t draw worth a damn, I got into to computers. I came to U Mass first to get a master’s degree in landscape architecture, and they had a dual-degree program, so I also got a master’s in regional planning. I was here from 1987 to 1991 for those degrees.
V1: I attended the American Planning Association meeting last year, and I was struck by the lack of GIS use in that community.
la Cour: Yes, it’s too bad. I was lucky, I joined a research group on campus that was looking at using GIS as a way to do landscape evaluation. I’ve always been an environmental preservationist, and the reason I came back to school is to get more skills to try to be more effective in helping protect our environment. I figured that learning how to design and plan would be the way to do that, and I saw GIS as an amazing technology. What’s exciting is that finally I feel that the technology is evolving to the point we’ve always dreamed about.
V1: Have you been following Jack Dangermond’s focus on GeoDesign?
la Cour: I was really excited to read about that. Jack Dangermond is a Landscape Architect and what I find interesting is how many landscape architects have had a huge impact on GIS. So much of it grew out of the Harvard Graduate School of Design back in the 60’s.
V1: Urban activities and 3D models are very interesting because it’s in the city where we can really increase our efficiency and reduce our impacts.
la Cour: Yes, absolutely. That’s why I’ve been pushing for 3D as a planning tool, not just for the information analysis, but also as a visualization tool. It’s amazing how if someone just sees a proposal for a new building and it’s not put in the context of everything around it in a 3D model, they have a very different impression of it. When you put that proposal in 3D, then they’re going to say, “Oh, my God, it’s too big, it’s way out of scale. Or conversly, “That’s OK, it fits.”
I convinced one of my interns when I was working for the Town of Amherst to do her master’s project on 3D GIS and 3D visualization. She did a block of our downtown, which we had already rendered in 3D, and did six different planning scenarios for it.
Right now, there is a historic cinema and a big parking lot, and a very new suburban style bank. There had been a proposal quite a number of years before for a three-and-a-half story building for mixed-used on the bank site, which would have been much more economically sustainable for the downtown. It got voted down because it was viewed as too large.
For one of the six different scenarios for downtown, I got the rendering for that building from the architect. It was by far the most intensive use of the site, but it was voted against in the town meeting. Now in 3D, it is their favorite scenario because they could see it at the appropriate scale. It wasn’t too big, it was actually the right scale. It fit in with the rest of the downtown fabric, and it would have been great to have the tax revenue from that building over all these years. 3D is a very powerful tool in helping people visualize future scenarios.
V1: So, 3D could have made a difference?
la Cour: 3D presents information in a format that people can understand. People have a very difficult time reading maps and plans, but if you can present the information in 3D or construct a physical model, they are quite effective.
It’s really nice what you can do on the computer now to create 3D views. If we’re going to change the way we do things, and try to become more sustainable and make things happen, you’ve got to help people see it so that they can buy into it, and support it in a public process. 3D is a really effective communication tool.
V1: Is your primary focus on building a detailed 3-D model for the university?
la Cour: That’s what we’re working on. When I worked as a planner for the Town, part of the master planning process was a 3D model of the downtown. Fortunately, one of the professors here at University of Massachusetts teaches an AutoCAD course and he created a bunch of SketchUp models in order to enter the Google Earth 3D campus contest, and has given us a good start for the campus.
What I’m looking to do now is to create a much more accurate and more complete 3D model of the campus, and hopefully have it done in time to play an important role in our master-planning process, which we’re just about to undertake. Our requests for proposals are out on the street and due next week for planning consultants to help us do our master plan. I’m hoping that by the time we really get geared up for that, that we’ll have all of our new data from a recent flyover, we’ll have our web-GIS up, and a 3-D model to work with.
V1: Does your vision for 3D includes floor plans and building interiors?
la Cour: Over the past few months, we have worked to take our floor plans from our buildings, get them geo-referenced, use our building footprints, digital frame model and our records drawing to provide base elevations. We very quickly extruded buildings and the floor plans in 3D based on a typical floor height and sandwich in between floors.
Because we have been involved in some academic master plans over the past year for several of our colleges, we wanted to be able to quickly provide a visualization tool to look at space. We extruded our floor plans and were able to attach these to our space database to keep track of all the rooms, hallways, and what department each room belongs to along with their use.
We wanted to create a tool to help people visualize where people were. Why does this department have one office over here in this part of the building and then the rest of the offices are down on another floor in another wing of the building? We wanted to help people see that spatial distribution of people rather than just looking at colored floor plans side by side.
3D certainly makes it easier to understand, so that’s something that we’ve been working on the last four or five months. Now that we did a quick and dirty model of it, we’re going back and using our record drawings to fill in floor height with the actual measurements to make it more accurate.
V1: Are there other efforts to add more realism to the 3D data?
la Cour: As part of our new flyover this spring, we secured LiDAR data for the whole town and the university. From that we’ve generated one-foot contours. I haven’t had the opportunity to work much with LiDAR before, but it blows my mind how much data is there. One of the things we’re going to try to do within the next four months is to extract very accurate wire frames for our buildings from that LiDAR data, and then export that into a SketchUp format for photo rendering of the buildings.
The idea is to continually add fidelity to our data, to go for the most accurate data that we can get. Our planimetric data is one inch to forty-foot scale, so plus or minus a foot accuracy. Now, with the LiDAR we’ve got one-foot contours. We also have surveyors on staff, so as they add survey data we’ll improve our accuracy.
One thing that blew my mind about the campus that I hadn’t realized was different than the town is that it’s a very dynamic place. It’s constantly changing, and so our surveyors are working hard to keep up. As new facilities come in, they survey it. So we’re also developing workflows to make sure that as new data comes in from the surveyors that we always use that most accurate source.
V1: I understand that you’ve been using Safe Software’s FME toolset to some extent in your 3D work.
la Cour: We were fortunate enough to visit Mike Parken at MIT, one of the great gurus of GIS in the facility planning world. He showed us what he had done with FME. I’m very fortunate to have a gentleman who is called, Alexander Stepanov, who works with me, and he is an incredibly sharp guy, who took that technology and ran with it and has figured out some very creative ways to use the extract, transform and load technologies to very quickly create a 3D model from our floor plans. We’re also looking at using FME to extract from our LiDAR data in order to create accurate wire frames of our buildings.
V1: As CAD and GIS converge, do you see using one tool over the other or do you think we’ll always be using both?
la Cour: That’s a reason I find FME so fascinating, because looking at our organization from a workflow standpoint everybody has got the tools they like to use and they create important data. As the GIS person, I want to bring all of that data together, but yet I don’t want to ask somebody to change the way they do their work and the tools they use.
For instance, the surveyor, has been learning GIS and has been working with it and, obviously, sees the value of the GIS, but his native platform is AutoCAD. What we’re finding is that with FME it’s a lot easier to let people use the tools they want to use and then be able to use that extract transform load technology to then bring it altogether in the GIS. You don’t have to disrupt the workflow to be able to use the different data that different people develop and use.
In terms of thinking about standards, one of the most troublesome spots about standards is that you so often have to dumb them down to be able to make them work. What I find really exciting about the extract, transform and load technologies, that in a lot of ways, it gets you around that. You can go from native format and put it in another native format and you don’t have to dumb it down because you have so much control about what you extract from this data source and how you transform it in the process.
It definitely is in stark contrast to the days when I first started out working at a landscape architecture firm, and I was the only GIS guy in there with a bunch of AutoCAD guys. We we’re trying to trade data back and forth, and it was a pretty cumbersome slow and not always successful process. You had to do the DXF out and DXF in and hope to get everything. That’s what I find so amazing about this tool -- the preciseness with which you can go into different data sets and extract only what you need, and then put it into where you want it to go.
V1: So, you have automated a lot of the workflow?
la Cour: Yes, and we’re looking at FME as a way to further automate that process, particularly with the LiDAR data. There are so many points that it’s a challenge to weave through the millions and millions of data points to get the ones that we really need. We’re working to explore how we can use these technologies to utilize the LiDAR data.
This technology helps streamline planning in terms of both providing the kind of data and reports that we needed to understand the situation faster than we could have any other way, as well by providing it in format that can help people understand it better.
V1: Are you addressing issues of sustainability as part of the modeling process?
la Cour: If you’re into sustainability, there’s some really powerful numbers in the forward to the National Building Information Modeling Standard that’s online. It talks about how the construction world hasn’t benefited a whole lot from new technology, and they haven’t realized the productivity gains that a lot of industries have through technology. One of the things they talk about is that buildings consume 40% of the natural resources in the world and emit 40% of the greenhouse gases. We have a huge responsibility to do a better job. The report indicates that by increasing the performance of a building by 3.8 percent you can save enough money to pay for the design, construction, and operation of the building throughout its lifecycle.
Three percent is not much of an efficiency mark to get a whole lot of gain. The other statistics that blew my mind was that in 2008 the construction industry was estimated to be a $1.2 trillion dollar industry in the United States, and they estimate 57% of that was non-value added or in other words, $600 billion dollars worth of waste. (I had to check the figures again)
Again, it’s important to fill in those black holes in the map, which are the buildings and creating these building information models to be able to be more efficient and sustain how we construct and operate buildings. Then putting that in the context of the GIS so that we can also bring to bear the good tools of site planning to improve how and where we construct them in the first place.
V1: We’ve been covering the BIM space and are really interested in the convergence of technologies, with GIS, CAD and BIM coming together, but there are a lot of headaches in making that transition.
la Cour: Yeah, nobody is quite certain yet how things are going to evolve. Although, I would say it’s not really a question of “if,” it’s a question of “when.” I mean you have the federal government and the general accounting office really pushing Building Information Models because the amount of efficiency to be gained is just mind-boggling. We’ve got to do a better job.
V1: So, would you say within the past few years, that we have leapt far forward?
la Cour: Yes, absolutely -- the horsepower in the hardware has improved to deal with the ever more complex models and the data is becoming much more commonplace, and the software is becoming more powerful and easier to use. We were amazed at how much more information we were able to collect this time around in the flyover.
Now, it’s time for us to have the people and the procedures catch up with the rest of it. I think that’s our biggest challenge. In fact, I was at a meeting this morning where we’ve formed a new committee to try to figure out how we manage all of the emerging enterprise technologies that are coming online, and how we get it all to work together.
V1: Are you creating a next generation of digital city modelers that will bring all these tools together?
la Cour: That’s certainly a key part of it. I think the key has been making the technology easy to use, so that it’s more intuitive for people to use because training is expensive and time-consuming. That’s the biggest trend I have seen in GIS over the last several years -- putting a lot more time and effort onto the back end of technology, so that the front end is a lot easier to use.
V1: What’s the ultimate dream of the 3D model?
la Cour: Ideally, the dream is to create a model of the world within the computer. The world is definitely 3D, and includes interiors and exteriors of buildings. That has been a fascinating thing for me coming to work for the university and entering the facility planning world -- it’s a lot about building. For most of my career in GIS the buildings have been sort of a black hole in the map. And we want to fill in those black holes because the buildings are mainly what we’re interested in -- building new buildings and maintaining the buildings that we have.
V1: It sounds like you’re well on the way to creating that for both the town and the university.
la Cour: Well, we hope so. I still feel like we still have a long ways to go, but now there is such a convergence of technologies that it’s very much possible now.
| < Prev | Next > |
|---|
