Data Exchange Unleashed: Revit, Rhino, and Power BI Integration in the Cloud

ALEX WOODHOUSE: Hi, this is Data Exchange Unleashed Revit, Rhino, and Power BI in the cloud. My name is Alex Woodhouse, and I'm joined here with Benjamin Doty. Here's the quick safe harbor statement. These statements are forward looking, and we at LMN and Autodesk take no liability for any decisions that are made from this presentation in the future.

So my name is Alex Woodhouse, and I'm a licensed architect and design technologist at TechStudio in Seattle, Washington, at LMN Architects. I'm involved throughout the firm's work across all phases guiding the firm in adoption of contemporary workflows and tools.

BENJAMIN DOTY: And Hi, my name is Ben. And I am a licensed architect. I've got 30 years of experience in the AEC industry. I've worked for architects, which I currently do at LMN Architects. I've also worked for owners and as part of the general contractor team. And currently, at LMN, I am the BIM strategist for the firm. So I touch and oversee all projects within the firm.

ALEX WOODHOUSE: So just a little bit about LMN Architects. We're 150 employees across architects, interior designers, urban designers, marketing, and administrative staff. We're one interdisciplinary office based in Seattle, Washington. Our market sectors include arts and culture, civic infrastructure, education, urban mixed use, and convention centers, and we're practicing throughout North America.

And lastly, we're using any and all software necessary to really deliver the work that we produce. So across the Autodesk AEC collection Rhino and Grasshopper, Adobe Creative Suite all the tools that really help us to communicate our design intent.

Additionally, the image you can see here is our new shop facility, satellite shop facility across the street from our current location. And as you'll see in this presentation, fabrication is a huge component to what we do as architects.

So as we look at our conversation here today, I think we really wanted to frame this discussion around how we can leverage data how can data redefine the relationship between design modeling and architectural delivery?

To quickly touch on the learning objectives, we're going to be learning about the fundamentals of Data Exchange. We're going to be learning about how to set up and configure Data Exchange files. We're going to look at it and examine a real world case study. And we're going to look at how this enhances our interdisciplinary collaboration in streamlining our design workflows.

So when we look at design models and architectural documentation, there's a few factors at play that are influencing the way that we're approaching this work. The first is that the traditional drawing set, really, as a 2D static representation of a project occurring at design milestones, those deliverables end up being out of date relatively quickly as the design progresses. So that's something we want to address.

Second, is the meaningful integration of BIM. So when I say that, BIM modeling is a newer workflow in the profession, by newer, I mean the last 20 years or so. And we haven't fully explored what it means to take advantage of the capabilities of BIM modeling.

Third, we'd like to understand how our parametric workflows and the increased complexity that comes with it, how we can leverage that information beyond just the design model but how it actually influences the documentation process.

Fourth, when we look at external collaborators, we're looking to how we can actually communicate information and coordinate with these external parties in more effective ways. And lastly, the evolving role of the architect in how our work can be delivered more effectively throughout the design process.

So when we look at design models here on the left, typically what we're doing is we're handing off information in a 2D manner, like, the drawing, you can see at the right. And the way you navigate a drawing set can be a bit cumbersome and at times, unintuitive.

In this example on the left, we see a particular location and detail identified. And that would require us to first, navigate to a floor plan where we're looking for a wall section. That wall section then takes us to a detail sheet, which ultimately takes us to the detail that we're looking for. So these examples we're going to look through are looking at alternative means by which we're using the design model to communicate information in a more direct manner.

So this first example you can see at left, is for an acoustic ceiling at a theater at the University of Iowa in Iowa City. And at the right, you can see these panel drawings that were being produced from the parametric model.

And the parametric drawing and output of this information really enabled us to more effectively and holistically document this design across thousands of components and all of the related criteria for each of those panels, all the different dimensions and perforation, IDs, and things of that nature.

The ability to work with this data allowed us to provide more quantitative criteria for the fabricators as they were initially reviewing this and both for their fabrication and bidding knowledge. This next project-- so the first one is about handing off information to the fabricator in the drawing set. This is actually looking at how the design model can be used or formatted for fabrication.

So rather than documenting as 2D drawings in the drawing set, this information was actually communicated as DXF deliverables to the fabricator. So at the left here, you can see a Grand Avenue pedestrian bridge in Everett, Washington. And at the right, you can see a variety of the different panel types that were exported as DXF files for the fabricator to then use their waterjet cutter with these drawings with these files.

So in this case, the relationship between the architect and the fabricator is much more intertwined, where the fabricators workstreams are actually informing the architect in how we deliver our design intent to them. The documentation method in this case was an exported file, but you'll see also that digital models can become the cleaner or more direct way to deliver this information.

In this next project, we're looking at the Octave9 Experimental Music Space in Seattle. And this is an expansion to the Benaroya Music Hall here in downtown Seattle. And in this case, the ceiling assembly was actually designed and fabricated here at LMN.

So the production of drawings for this, which you can see at the right, were instead less of something that was being communicated directly to a fabricator but actually, a working process for us to then use as we were fabricating CNC cutting these fabric felt panels.

So in this case, the architect is the fabricator. And mockups and prototyping are really integral to how we, not only designed this, but ultimately, at the end of the day, how we were able to fabricate and assemble and install these components in the finished space.

In this case, the inherent design knowledge that we were using was a bit more unique in terms of how we needed to communicate it being that we were fabricating it ourselves. And so there are some nuances to the delivery process that in this case, we were able to expedite not having a third party fabricator as part of the project.

So another example where the design model can be effective, instead of the fabrication process, is actually just providing more clarity and more definition to the geometry. So in the case of the wood cladding at the Ocean Pavilion, also here in Seattle, which is currently in construction, the panels themselves couldn't accurately be communicated as 2D elevation drawings.

We found that there were drawings in association with data tables, and there was a disconnect between the text information and the graphics. And it actually became much more nimble and intuitive to look at a 3D model and have dimensions tethered to that so that we could actually understand the geometry and the out-of-plane geometry in real space.

So in this example, the 3D models are much better at visually communicating that geometric or volumetric variation. The data tables and drawings tended to be unintuitive and not really the best way to represent this information. And in the case of a tool like Revit, the BIM model was not as effective at communicating the deviation from one panel to the next because of the project-based location. Whereas, here, as you can see in this GIF, each panel being overlaid atop one another allows you to really understand the difference in plane that's occurring from one panel to the next.

So next, looking at the design model, really, as the only way to get a firm understanding of what the geometric complexity is about a particular element. In this case, also at the Ocean Pavilion, we're looking at the coral canyon tank, which is a 300,000 gallon concrete assembly. And in this case, this geometry could not reasonably be communicated through traditional plans and section drawings.

So we were actually using the entire model as the deliverable to the contractor and subcontractors as the primary means of communication. So this includes the general contractor, as well as the concrete contractors, the rebar contractors, the formwork for the concrete, the rock work subcontractors, the acrylic manufacturers. All of these parties were referencing one digital model for the production of their particular aspects of this project. So a lot coming together through a 3D model.

And lastly, an example here using the design model as a real-time representation of an element. In this case, we're looking at the Forest Trailhead project here in Seattle at the Woodland Park Zoo and particularly, at the ceiling assembly, which is a mass timber construction.

So as this design has developed, we're actually outputting these drawings you see at the right, in a real-time representation of the actual quantification of parts of different beam lengths of the quantities of these beams of the profiles that is being handed off directly to the contractor for estimating and coordination purposes. In this case, we're actually not waiting for a milestone, but we're instead giving this to the contractor in a recurring schedule so they have real-time understanding of what this design is entailing.

So we've looked at six different examples of current and past projects in a variety of ways for how the design model is being used beyond its traditional means as a design tool but for clarification of design intent. And what we'd like to talk to you guys today is about how maybe this might evolve in a next iteration looking forward with the Data Exchange tools.

So if we think about a design model as an interactive dashboard, maybe this is a tool or an interface that's bundling the model geometry, as well as two-dimensional drawings and the corresponding metadata that we have access to through BIM modeling.

The modeling platforms and proficiencies vary across disciplines. Contractors are using different design and model software than architects, and that might be different than a structural engineer. So having something that is intuitive and platform agnostic, promotes inclusivity and understanding across all parties and disciplines.

And lastly, what's the most effective way to share information? So in some cases, certain design parties might not need the same level of detail or scope or breadth of the model, and an interactive tool might actually provide the ability to filter, refine, and otherwise better control the information that they're viewing.

So with that, this idea is going to be great as a concept, but we'd like to get into what an actual case study might look like. So at this time, I'm going to hand off to my colleague, Ben.

BENJAMIN DOTY: So what we're going to highlight today are the various data connectors that Autodesk has to offer us. These connectors, basically, allow interoperability through-- or connections to various other programs. So in what we're seeing here, is that the data connector can connect information between Revit, Rhino, AutoCAD, Inventor, Civil 3D, Dynamo, the Power Platform, or Power Automate, and we will be looking at Power BI.

And what this really allows us to do is that different programs really do different things better. Rhino is a great, great modeling platform. Revit, meh, maybe not so much, but Revit is a great database. And while it doesn't necessarily display data in a really great way, Power BI does. And these data connectors allow us to look at these, or allow us to connect these different pieces of geometry and data together and allow us to let those programs do the things that they do best.

And so if I go over next. And so what we're going to be looking at is this rainscreen cladding for a project that we've been working on. It has these 1,400 plus individual panels that were all done via, or created via Rhino and Grasshopper. And then once those were created, we brought them into Revit, populated them with additional metadata, and then were able to then bring these into Power BI to create a dashboard for all of the elements. Let's see here.

And so this is looking at the geometry inside of Grasshopper and what came out. And so we have these data connectors, and so the easiest data connector would be directly from Rhino and Grasshopper into Power BI. However, Rhino is not a BIM product.

And so it does not have an inherent data structure that you can actually analyze in something like Power BI. So while it was great and the geometry comes through, the actual robustness of the information was very lacking. And so we started looking at other options, and we started looking at Rhino Inside, and then using the data connector into Power BI.

This was really great. We used direct shapes as generic models. Very light, but again, with direct shapes, there's just not a lot of data that you can really harvest from those. And it's hard to put new data into those elements.

We had more success doing the same process with adaptive components, and that was actually really great. The adaptive components came in as generic models or generic model elements. So there was a robust set of parameters that were already in place that we could use and start to interrogate the models, once we get them inside of Power BI.

But again, because ultimately, this is a Autodesk University project, we worked with using the data connector from Rhino into Revit and then a data connector, again, once that information was populated or additional data was added into the Revit model, we used the data connector, again, to go directly into Power BI.

And so that's sort of the workflow here. So this is the Rhino model. As we saw earlier, the data connector, you enter in the command. It's command line. And it's very simple to set up a new data connector. And there we go. Create a new connection, add a new name or create a name for it.

And this process is very quick. So I think these 1,400 panels, I think, it took less than two minutes for it to translate. And so then, once we are able to translate it from Rhino, this is the Revit side. And so I say, Update, give me the information. And within 30 seconds, we have that Rhino information into Revit.

And so at this point, we can start adding type-based parameters and instance-based parameters. And so that's really kind of an important thing, I think, to understand that you get both levels. And you can add them-- once it's inside of Revit, you can add that information through project parameters. It doesn't need to be shared, but it can be. But it's very, very, very powerful to be able to add that additional information.

And so then once we are in Revit and select our information, or select the elements we want to export, and this is the Data Exchange process, very similar from the Rhino side. Give it a new name, provide it a location.

In this situation, everything is going into the Autodesk Construction Cloud. We have a dedicated site for this. This is using Revit 2023. I believe it's 2023 and 2024 can use the data connections. And then at this point, it gets loaded into the cloud, which it did previously.

But then once that happens, we can then come into Power BI. And on the left here, we can see there there's a 3D visual that gives us all the 3D information. And because Power BI is really, really good at looking at data, all the graphics and the visuals are interactive.

You select on one, it displays across all of the others. And in this example, we have brought everything in as the single category as generic models. So we see everything as one color or one yeah-- shown in one way.

When we add a little bit more information then to the dashboard, we can start to color code the individual panels based on their type and their location. So again, you can see, by looking at the visual, select one visual, and it highlights and then cross filters through the other panels.

Incredibly, powerful, and then this is looking again, I talked about already, instance versus type-based properties. So we're looking at type properties in this example. But in the next, we can start to look then towards instance base properties.

And so for this example, we said that there were five different types of finishes for these panels. And each of those were randomly assigned via Dynamo. And then we can see that there's the different types of panels in different finishes. And they all sort. And you can see on the upper right, that those are then accounted for depending upon the individual panel that's selected. And again, Power BI does really well. It displays data in a very clear and concise way.

So what we then looked at next, was starting to look at individual panels and their location. And so this would be more of something of a field location. So if I clicked on an individual panel on that 3D view, . It gives the panel location. And you can see above that, the panel_10-AH-23. So that gives a horizontal and vertical location. And then also you can see that there is the panel cut information that would be provided as a DXF or other file format that the installer might need to cut these.

And then there could be a QR code that would be assigned to each panel. And so if out in the field, you could then scan the QR code. And you'd know exactly where that panel was going to be placed.

This whole process is really-- as we've said, this is really a case study. We're still trying to figure out how to actually implement this type of tool or this workflow. But I will say that Autodesk has been a great partner on this while we have been working through this.

We are very much in the beta or alpha and beta phase. But I would say, on a weekly basis, Autodesk has been providing us updates and trying to understand what the use case would be for this. And so they've been a really great partner in this.

And so but as the lessons learned, the previous slides show that there's a lot of different ways to bring data from, in this instance, Rhino into Power BI through intermediary steps. And I think that's been a challenge. And trying to figure out how that actually is going to work for us in real life.

The Power BI model visualization tools for 3D elements are pretty limited. I think Autodesk has done a good job of incorporating their tools into that. But I think it's really what I've-- really my takeaway on this is that it's on a case-by-case situation. For what we were looking at today, this was the best situation for us.

If looking at other different types of delivery elements, I think there would be potentially a different way to look at it there. So I don't think it's a one size fits all is really what I'm trying to go for.

ALEX WOODHOUSE: Great, thank you, Ben. So we looked at one case study. But just to look at a few additional applications of how this might be useful, we have some ideas as different possible dashboard interfaces.

The first, which you can see here at the right, would be a structural element viewer as a supplement to the structural framing bid package, where a prospective subcontractor could receive this document and have an easily navigable format for viewing different model elements that they'd be bidding on as part of their bid package.

Another example would be potentially an FF&E visualization tool, either for the client for approval processes or for estimating reference, again, as budgets are coming together. A third option could be using this as actually an interface to view different design options and having quantitative and cost information to react to in addition to those design options.

And lastly, this could be a tool, as Ben was just mentioning, an on-site reference for element coordination and installation, where you're actively using, potentially, a QR code to reference and scan actual parts and have drawing references available on demand.

So as we wrap up this case study, we'd like to think about what the future might hold and how these ideas can continue to evolve. What does the future of architectural delivery look like? And how important is it to have access to this portable, real time, and interactive 3D data? How does the delivery of architecture change as our tools continue to evolve around us?

And with that, we'd like to thank you for joining us today and hope that you learned something about the Data Exchange process and the delivery of architecture. Thank you.

BENJAMIN DOTY: Thank you.