The Inconvenient Truth About Copy/Paste and BIM
(AKA: “Why can’t I simply copy and reuse the design element I figured out in my last project”)
With Skema, a firm's design standards stay intact, because we start with their Revit data in the first place.
Architecture firms have amassed a huge amount of BIM models and data over many years. And yet the ability to tap into that repository has been limited by technical obstacles. Until now. We are now reaching a point where those obstacles are being overcome, and a time when firms can apply technology to make a big leap in efficiency and productivity.
To explain, first I’ll do deep dive into those technical obstacles, specifically as relates to Revit.
I describe Revit as a cardboard model in a building that knows how to draw your construction drawings. All the construction drawings are views of the model with annotations and dimensions put on top of it. They’re connected. There's no way to have uncoordinated construction drawings in Revit. If you make changes to the cardboard model, the construction drawings reflect it instantaneously.
That cardboard model is stretchable, parametric, which is how it's able to do what it does. But a downside of that, especially as a Windows desktop application, is that Windows applications have to support Copy-Paste.
If you've ever tried to copy and paste any amount of data out of Revit into a blank project (and I have), you realize that's really difficult. Because what happens with the parametric nature of Revit is that walls are attached to other walls, and floors are attached to walls, and ceilings are attached to walls, and walls are attached to roofs, etcetera. There's a huge map of nested, interdependent things.
When you copy data out of Revit and try and paste it into a blank project, you run into major problems, because it can't remap all the references. Each wall is looking for something to join to and there's nothing. And that has made it really hard for people to reuse what essentially is their intellectual property.
Of course, there are workarounds around that, but most people don't get to the level of planning needed upfront in a project to be able to pull that off. A brief shout out to computational designers. What they do is amazing and I have great admiration for them. But computational designers in a firm are a separate breed, and most designers don’t have that skillset. Nor should they have to, to be successful architects.
We asked ourselves, How can we make Skema be easily used by anyone in the firm, even studio principals doing hand sketches?
We believe we have developed a fix by applying the deep knowledge and expertise our team has of Revit, coupled with an understanding of what issues frustrate and slow down a design team when producing BIM deliverables.
And the idea of this fix is based on a book that was written 20 years ago called Refabricating Architecture. The lessons in the book are still relevant today, even as we go from physical construction to virtual design.
The book describes the physical difference between construction and manufacturing, and it drew lessons from manufacturing for the construction industry.
In the manufacturing industries, the manufacturers create chunks of things offsite. You can call these chunks, components, or modules. Then the manufacturers make standardized interfaces between them so they can then bring them to a final assembly area. The Airbus plant in Lyon, France, for example is where all of the chunks of an airplane are brought together to do final assembly of planes. And the chunks all fit together perfectly because the interfaces are specified and made in a modular way.
Today’s modern methods of construction and prefab and panelization are all implementations of that idea -- physical implementations. However, unlike manufacturing, our design software’s capabilities have not yet caught up with the industry’s desire to replicate manufacturing methods.
SKEMA is designed from the ground up to support these ideas. We're taking chunks of projects in Revit and reusing them, and we solve the building in chunks rather than at the level of wall-door-window. So instead of dealing with wall-door-window in SKEMA, you deal with whole blocks of the building. And then because the chunks are deterministic, we can generate the building in Revit from whatever arrangement you make.
So, I give you a term I’ve been using: meta BIM. It’s a higher level of abstraction. We're not working with objects, we're working with chunks.
What are the implications? One of the things which is really different about Skema by design is this idea that we call a BIM knowledge reuse engine. We are not just making a mass, then chopping up in block and stack and creating several options, and then redrawing the best option in Revit, where additional design work occurs. Redrawing in Revit is inefficient.
Instead, we asked the question: Why can’t we take chunks of buildings out of Revit and use them as our starting point for our design?
And that’s what we’ve created with Skema.
Let me be clear: our goal is not to make a hundred percent finished building. Our goal is to take the repetitive part of the building, let’s say that 60% or 70% of the building, and automate the design of that and leave you to do the remaining 30% or 40% of the building, which represents the interesting parts, and the parts that matter, like lobbies and experiential parts of the building.
We've been holding projects charettes with architects and generally, we'll work on a design over an hour and a half or so, and then we'll push a button, and in a minute we'll get the finished BIM model. And then we ask the questions:
How long did it take you?
How long would it take you to do all of that by hand in Revit?
And the general consensus, at least on residential projects, has been about two months. That’s a two month fast-forward button. The implications are immense.
And there’s more to come. We have some major updates coming that expand the typologies that Skema can address. I look forward to sharing that amazing new functionality with you.