Autodesk Charts a Path to Enable More Integrated and Sustainable Projects

Autodesk has been pulling together a portfolio of products around the efficiency gains that can be realized with model-based design. The product strategy of their AEC group has evolved to specifically focus on project design, performance and execution now, rather than simply individual design and engineering products. V1 editor Matt Ball recently attended the Autodesk AEC Technology Day in Waltham, Mass., and spoke with Phil Bernstein, vice president of strategy and industry relations, about this move toward integrated design.

V1: I was pleased to receive a copy of Building (in) the Future, the book that you edited for Princeton Architectural Press. It brings together a good deal of thinking about sustainable design and the role of model-based design, including the changes needed in the workforce. The role of design has been such an individual pursuit that I’m wondering about the steps needed for a more collaborative process.

 Bernstein: It took Princeton some time to pull the book project together since the original symposium behind the book took place in 2006. Because things are moving so quickly, I was a bit worried that we would have a relevance problem.  As it turns out, a lot of the content from the contributors anticipated a lot of the things that we’re thinking about right now.

Autodesk is very focused on understanding the relationship between all of our technology pieces. We had to go through a period of optimizing each of the pieces for their own usefulness, and now we’re starting to look at the relationship between BIM and GIS, and how to link our geospatial tools with our building modeling tools. We need to look comprehensively at the sustainability of a project, instead of the sustainable characteristics of a data set for example.

This is what our contributors talked about as being the problem. The nature of the work process in a project changes by the very nature of this digital technology. Traditional barriers between processes are eliminated. In the book, Paolo Tombesi talks about the idea of flexible specialization, where the design process can no longer be segregated by labor topologies. Ron Merchant discusses how his team did a comprehensive supply chain integration of steel frame structures. It’s all about the integration of the problem, rather than its individual pieces.

V1: The idea of driving down the time, money and resources of a project, while also doing a better job of reducing impacts on the environment is a large part of Autodesk’s message. Do you equate sustainability closely with efficiency?

Bernstein: Efficiency is largely a part of the effects of sustainability, but for us a lot of it starts as an ethical obligation. We’ll ask ourselves “What’s the right thing to be empowering our customers to do?” We have millions of designers that we empower with our tools. We want to encourage a more sustainable attitude as part of their interaction with our tools. It’s not just about the business opportunities with our technologies, it’s the right thing to do.

Also sustainability is about looking at projects in a comprehensive way. It’s not about electrical drawings, roadway drawings and site drawings; it’s about the understanding of this organic thing, and the responsible ways in which to make it operate in an integrated fashion.

This word integrated keeps coming up in a lot of different contexts, whether it’s the delivery of our Trapelo Road headquartersor the idea of flexible specialization that Paolo talked about, or our whole sustainability strategy. It’s all part of the same theme, and it seems to be the theme of my existence at the moment.

I’m currently reading a book about the relationship between spirituality and science (Einstein’s God, by Krista Tippet). The chapter that I’m reading right now is about a cardiac surgeon at Harvard who runs this alternative integrated medicine practiceHe combines the traditional cardiovascular structural surgery with reflexology and holistic medicine. He emphasizes the need for a global perspective to understand things, and to start looking at the adjacencies and not think that all problems are about fixing heart valves or whatever. Integration is a pervasive social issue right now.

V1: I was pleased to sit in on an Autodesk Labs presentation at this year’s AU Conference where algorithms to inform design was discussed. I was really fascinated by that, but I can imagine that from a designer’s perspective that it’s almost too much of a role for tools. What is the role of technology in the move toward thinking more holistically?

Bernstein: There will be initial intimidation followed by the increasing realization that a lot of the more banal aspects of the design problem can be parameterized and solved algorithmically, freeing a designer to work on the more interesting stuff. We’re looking at some technologies around design optimization, where essentially if you can describe the design problem parametrically, you can algorithmically limit the number of potential options so that you’re not investigating options that don’t work. What you’re then able to do is define the space in which a solution can reasonably work, and therefore more systematically investigate what the options are.

That’s where we’re going. Design will always be this highly synthetic and intuitive thing that you can’t do entirely algorithmically. You’re not going to be able to get to beauty algorithmically or you’re not going to get insight and see things in a new way algorithmically. What you can do is program a computer to address issues such as the energy efficiency of a building, the number of pounds of steel needed, lighting levels, etc. As a designer, that’s what I want. I want the computer to tell me that the reason the light fixture works is that it spreads a certain amount of light over a certain area. That way I can worry about positioning that in a way that makes it look cool. You’re starting to see both in academia and in our own work, a lot of theory on how design optimization might happen.

They do a lot of this computation-intensive design in the aerospace design industry. They might be looking at the design of a new wing, and they’ll write a huge number of formulas to constrain the answer space so that you’re only exploring the ideas that will work. There are a wide realm of solutions that work, but at least you’re not wasting your time working outside the boundaries.

In a building project, there are 50 or 100 issues that would be great to solve algorithmically. I’d like to be able to some day apply algorithms to a building information model and have it do all of the very boring things like, “are the fire corridors wide enough, is the stair wide enough, are the doors all swinging in the right direction.” We should be able to do all that kind of tedious work algorithmically.

We probably can’t do that until we’ve arrived at a robust-enough representation of the design. If someone is working on a drawing, it’s very hard for it to have enough insight into how a stair works. A parametric model of a stair can be adjusted to say the population of this floor at 116 people, therefore we need 16.2 exit widths, therefore the width of the stair needs to be 83.27 inches wide. That’s just arithmetic, and it isn’t even interesting arithmetic. If we have a representation of a building we can do these things.

V1: I’m also interested in the nexus between the built world and the environment that considers all inputs. There is so much that we don’t know about our impacts on the world around us. Is Autodesk going in that direction as well in terms of trying to tune designs to the environment?

Bernstein: We just finished updating the weather database that was available, increasing the number of data points to a precision of less than 8 miles, where it used to be 25 or 30 miles. The weather data that we’d use for the design of this building we would get from Logan Airport, which is 15 or 16 miles away, on the water, and next to a city. We just upgraded the weather database so that you can get the data for local areas for the whole United States. It’s a first step toward a level of granularity of data to be able to handle local problems.

When one of our customers geolocates a design, they’re starting to get much more insight into how that building relates to the environment around it. You can move the sun over the building with a great deal of precision, we can model the energy, electricity, heating, cooling, and we can do the weather simulations at a much greater level of precision. That is just going to continue.

V1: I was talking to an engineering company that is designing a storm water plan for a mixed-use development. The old way would dictate a long stretch of pipe to discharge that water directly into the river near the site. The new approach for low-impact design is a much more intensive and on-site interaction that requires a whole new approach. Do you feel that these new approaches are guiding your own tool development?

Bernstein: Absolutely, because if you’re drawing something, you’re using a spreadsheet and calculator to understand how much water there is, and how big of a pipe you need to draw with two parallel lines between my project and the river. It’s a very reductive process. If you’re using a model, you can begin to simulate your design and explore alternatives much more rapidly, which allows you to try a number of low-impact scenarios because it’s relatively straightforward. And our tools will tell you rather quickly how much water, where it comes from and where it goes. You can start to try different alternatives, which in a paper-based environment is really hard to do unless you’re getting paid specifically for that.

We think that a fundamental characteristic of sustainable design strategies is working from models instead of from paper. A simulation is a more robust platform to do sustainable design than a diagram or abstraction.

V1: You straddle both academia and business. I’ve seen a number of new departments crop up in the earth and environmental sciences in the academic world that take a multidisciplinary approach to a lot of broad problems. Is that kind of thing beginning to happen in the design space as well, where planners and architects are collaborating more?

Bernstein: Where I teach at Yale, we run a very small, very design-oriented graduate school. We have 180 students in our entire program. We only grant two joint degrees. One is a joint master of architecture and MBA, and most of those people become developers. The other is a joint masters of architecture and forestry, which is where the environmental movement in the United States began. It’s basically a green masters of architecture, with half of your classes in architecture and half in forestry where you’re trying to bridge the two programs.

I teach professional practice, which is business ethics. I’m toying with the idea of going to teach my same course, but in the forestry school, exploring what it means to describe practice and execution when your fundamental motivations are sustainability and not traditional building execution. In my copious free time, I’ll design and teach that course. [Laughter] It’s all completely about these crossovers.

V1: You’ve talked about the collaboration and multidisciplinary approach within the toolset, with that functionality being extended. Are there pieces missing?

Bernstein: There are a lot of pieces missing, but it’s really happening on two axes. One axis is the degree of collaborative transparency that a model-based tool gives you, even if disconnected with another model-based tool. If a design team sees a problem in a Civil 3D or geospatial model, they can address the problem much more clearly. The other axis is the direct connectivity that we’re starting to work on in our tools so that you can put a Revit building in a Civil3D model, and that you can locate a Civil3D model in a Topobase model. We’re starting to work on those adjacencies.

V1: I really like the analytical capability of the Green Building Studio, and the idea of using the cloud to analyze models. You also have Ecotect for building analysis. Is the approach now more of a cloud-based analysis as a service?

Bernstein: We’ve merged Ecotect and Green Building Studio and are putting some of the functionality of Ecotect into the core of Revit. Just yesterday we announced that users on subscription are going to be able to access our Green Building Studio on the cloud, and do a lot of calculations transparently. We’re working to make it literally push-button analysis. You can’t easily do that level of analysis on the desktop, because it might be four days until you’ve reached the answer. By pushing that analysis to the cloud, with a couple of thousands CPUs, you get your answer very quickly.

As an end user, you really don’t care where the functionality lies as long as you can access and use it. It’s like my daughter who is a student who functions completely on a netbook. She keeps no files locally, she keeps it all on the cloud. That’s the way things are moving.