Autodesk Forge Data Road Map: Moving from Files to Cloud Data

FARZAD TOWHIDI: Hi, everyone. I'm Farzad Towhidi, product manager on Forge Data, and I'm here with Philippe Videau, who's a senior product manager on Forge data. And today, we're going to be talking to you guys about the Forge Data roadmap and Autodesk journey moving from files to Cloud data. Next slide, please. But first, we'll start off with a safe harbor paragraph which I assume everybody just read. But if you haven't, it's just a reminder that the statements made during this presentation reflects our current expectations and assumptions and that no purchasing decisions should solely be made upon reliance on these statements.

So we'll kick things off by first describing some of the industry trends that we're seeing and customer challenges that our customers are having, and we'll go over two solutions that Autodesk is working on as an answer to these customer challenges, first one being the cloud information models which bring data, granularized data, to the cloud, and data exchanges, which allow you to share seamlessly your data across apps and teams. And then we'll end things with key takeaways and how you can get started on these two solutions.

But first, let's kind of talk about some of the key trends that we're seeing. So throughout the last few years, we have observed the following three trends that have been emerging and are now accelerating due to the pandemic. Next slide. Yeah.

Our customers need more automation, whether this is to increase productivity, improve efficiency, better resource management, or just faster time to market. To paraphrase one of our customers, manually synchronizing CAD and ERP data by opening files to extract basic CAD information and entering them in the ERP systems wastes a lot of time of our employees. Automating this process will greatly improve the efficiency of our workflows so that it frees up time to be spent elsewhere. Next slide, please.

And we've also seen increasing trends towards industry convergence. For instance, we've seen workflows that converge both AEC and the media and entertainment industries through the utilization of AR/VR technologies for reviewing and coordination workflows. Slide.

And there is a burning need for increased collaboration whether that is within or across disciplines and organizations. Since more and more customers are working remotely, which has been accelerated due to the pandemic, our customers are forced to create these bespoke data sharing tools that are non-differentiating to their business and that are also expensive to maintain. The movement of data between applications becomes a critical foundation for effective collaboration.

So working with files is painful. There is no standard way of collaborating, which you as Autodesk's customer are left to build bespoke collaboration pipelines using many different tools to get your job done. And also, design and make files are big and bulky, which takes a lot of time to transfer, translate, and store. And your projects generate hundreds if not thousands of files with lots of versions to reconcile.

Without asking someone, it is really challenging to get the right version to work on. Next slide. It's quite apparent that we need better data design to make data management. We need to ensure that applications are interoperate so that you can focus on your project and not building infrastructure to move your data around. We need to accelerate your workflows by providing efficient data access and manipulation so that partial reads and writes are lightweight operations to the entire design and increase your productivity by providing the right data available at the right time to the right people.

And this is done through emerging technologies that we call cloud information models and data exchanges. So let's start with crowd information models. The fundamental idea with this new technology is to break up files into different aspects of the design and make data and manage this in an open and transparent data model so that we reduce the complexity of applications working together by providing a single source of truth that all apps can access and count on.

We increase your productivity by manipulating and accessing only bits of data, and we can make sure that users can trust the data they are working with because every change for every granular bit is versioned so that you are always working with the right data. So our vision at Autodesk is to provide a cloud information model for all industries that we serve, whether it's in the media and entertainment, manufacturing, or AEC space.

And today, we have a cloud information model in production for the manufacturing space called Fusion Data, and we are working on a similar cloud information model for the AEC space this year called the AEC information model. Let's look at how some customers are using Fusion data today to provide value for their end users. Fusion data enables non-designer personas to be able to view and interact with CAD data outside of the authoring tool.

Without this capability, the view that you see on screen is costly to generate outside of the Fusion 360 app because it requires a translation step on the F3D file. And with the help of Fusion Data APIs, companies can automate tedious and repetitive CAD data import processes into other systems. In this case, a customer is automating the process of synchronizing their CAD and ERP data which, prior to these APIs, they were forced to open each design file and manually enter the CAD information into their ERP system.

And running calculations off of the CAD model without the authoring tool becomes a possibility thanks to Fusion data. What you see on screen is an example where a customer is calculating the environmental performance of their manufactured part as the CAD model is updated without any manual data entry, which is the norm of how the environmental performance is computed today. And as you may have noticed, in all three examples, the CAD data was accessible without the need of writing an application plug-in. That is the true power of our cloud information model.

So you just saw a glimpse of Fusion data in action. Now let's go over the capabilities available for you today. With Fusion data, you are able to extract granular data from your Fusion model through direct access to parts of the models like thumbnails, item number, materials, and other properties. Fusion data is also a source of truth for the CAD bill of materials. Our APIs allow you to download or traverse the most up-to-date hierarchy of parts within your model. You can also receive webhook notifications when a new milestone has been created and trigger downstream workflows.

And Fusion Data is your One stop shop for project and CAD data, allowing you to traverse the hub, project, folder, file all the way down to the granular CAD information [? with ?] a set of consistent APIs. And these APIs are powered by GraphQL, a bleeding edge technology in the Cloud API space that are simple to use and human readable.

So now that you have a high level idea of what you can do with Fusion data, give it a try by visiting our website and following our Getting Started tutorial. So we just went over one cloud information model that is available today, but we are also working on an information model for the AEC space. Let's walk through two sample workflows that AEC Sim will be able-- that the AEC cloud information model will be able to power in the near future.

So the first example workflow demonstrates a column schedule web app that allows contractors and structural engineers to generate and view an up-to-date column schedule of a structural model outside of Revit so that they can access this information from tablets and low end laptops. The workflow starts with a contractor or the end user that logs into a third party website application which provides an up-to-date view of the column schedule for the latest published model. In this application, the contractor can filter the columns based on the specific type of column and other column properties, and they can select a single column and view a longer list of detail properties. And all of these would be powered by the AEC cloud information model.

Now, the second example illustrates a web application that calculates the embodied carbon of a building. In this workflow, every time the design is published, a third party backend server is notified and extracts all of the elements made of steel and concrete, which is then used to calculate the embodied carbon of all those elements using a carbon calculator that might be available out there. So these two workflows were just a subset of potential workflows that the AEC sim will enable.

And so by now, you are probably excited to know more about what's to come. Here's a glimpse of our cloud information model roadmap. As we gather feedback from our customers in crystallizing our vision, we have mapped certain key categories of capabilities or data that we will surface in a cloud information model in a mid-term roadmap. I will not go over each line item but wanted to call out the items that we are working on now.

So in the Fusion data swim lane, we are working on allowing you to extend the Fusion data cloud information model with your own custom properties and also surfacing lifecycle data as well as physical properties. And on the AEC information model, we are working on exposing the published building properties from Revit. Now I will hand off the mic to Philippe to talk about data exchanges.

PHILIPPE VIDEAU: Thanks, Farzad, for that portion. We're going to jump into data exchanges, which is another critical piece of solving some of your collaboration challenges, making sure, like Farzad said, the right data ends up in the hands of the right folks in the tools that they're comfortable using with-- that they're comfortable using. You know, it's one thing to have the source of truth of your design and make data in the cloud, but how do you collaborate only on the data you need, and how do you make sure that that's a seamless and secure experience across the board?

That's where data exchanges come in. In the next few slides, I'll dig into data exchanges how you can start to use them, if you're an architect, engineer, fabricator, or end user of our design applications. Also, if you're a developer too, we've got APIs and kits that you can use to start to build integrations with data exchanges.

So we'll go through all those, a couple of demos, our roadmap and then what's coming up. So like Farzad said, as we move from app and file silos to transparent cloud data, it's important we provide collaboration tools that support managed, trackable data sharing workflows. That's exactly what exchanges are targeted at doing.

So customers can share the right data with the right person in the right context instead of sharing massive files, massive models, with minimal control over data noise and who has access to what. So as this diagram shows, sharers of data can protect IP while downstream receivers selectively consume the data they need and they care about in their app of choice. All the while, we are reducing latency across these collaboration workflows because we're just sharing the bits of data that customers need, not massive 100 megabyte files, gigabyte files in many cases. So things can happen, collaboration can happen, a lot faster.

So let's jump in to how data exchanges work today, and we'll start by looking at the end user perspective. We're going to kick it off with a pretty common customer example that we hear a lot about. And of course, there are hundreds, thousands of very, very similar collaboration examples. But let's say we've got two individuals here working at different companies-- an architect who's working on a building, potentially an office or a hospital space.

And we've got a stair railing fabricator who is focused on designing and manufacturing stair railings. Now, in many cases, many similar scenarios, the architect, the author of the design data, doesn't want to share his whole model with the downstream consumer like this fabricator. Maybe it's to protect the IP, certain parts of the building.

Maybe he doesn't want to overwhelm the stair railing fabricator with too much data. Now, on the fabrication side, our stair railing fabricator, she really only wants to deal with the data she needs as well. And more often than not, that includes a subset of the model rather than the whole model itself.

So in this example, the architect with data exchanges can select a subset of data to share with the fabricator. That subset is what we call an exchange. It can be designated by app-specific filtering mechanisms. In this case, the author would go and use a 3D view to specify the exchange itself. That exchange gets created. It can be created with one of our in-app connectors on the Revit side, or it can be created by users who have access to Autodesk Docs, which is part of the Autodesk Construction Cloud.

Now, once the exchange is created in any of those tools, it creates that subset of data, that secure subset of data. Now, as opposed to models and files, an exchange is actually stored in a cloud database. Each, every object here, its associated properties, the relationships between those objects are stored granularly in a cloud database rather than a blob file on the cloud.

Now, with that exchange created, the author or whoever has permission to share that exchange can go ahead and do that to the various stakeholders in the project. So in this case, that exchange is shared with the fabricator. The fabricator can open up Inventor and, with the integration we have with Inventor, can bring in just the data she needs. So in this case, we're bringing in the geometry data and some of the family data that you see on the left hand hierarchy tree side.

Now, this is just one of the integrations we built. We've built many more, which I'll get into. But one of the key aspects of data exchanges in many of these collaborative workflows is every person who has access to that exchange might want to receive updates when that source model is updated itself.

So here, we're just showing a cosmetic change where the author or the architect here updates the Revit model. Those updates can be automatically or manually pushed to the Data Exchange. And then the consumers of that data, in this case an inventor, can pull that data in when they desire. Now like I mentioned, we have other applications that we're integrating data exchanges with.

It's not just Autodesk applications. We're building out integrations with a tool that you see in the bottom right, which will have a video on, Microsoft Power Automate, a great workflow automation tool where you can pull parameters from the data exchange here to power reports and dashboards. We're building a integration with Rhino 3D, which is used in many conceptual design workflows, especially with Revit-- so looking at integrating with those apps, which are non-Autodesk tools.

And then for developers, our Cloud APIs are open. They're in public beta today. You can go and use those along with a set of geometry SDK to read in property data-- that rich, granular property data that I talked about earlier-- in that exchange as well as the geometry in that exchange.

So let's get into some demos that really show off how data exchanges are being used by some of our customers and early partners. I'll start with the Revit to Inventor workflow, which is the example I just showed. So I'll just go ahead and play that video.

Now, one thing I'll get to is right back. So I'll talk about at the end of this video. But here, you can see the architect is just designating a set of 3D views that he might want to create an exchange from.

Here, we're publishing the model to Autodesk Construction Cloud. And once that model is saved and processed, again, we can create an exchange directly from any of those 3D views and save it to a folder which our stakeholders have access to. Now, once that exchange is created, our Inventor user can go in and access all the exchanges she has access to in a project, load that exchange in, and now she can start to design that stair railing fabrication around the geometry and some of the family hierarchical information that she gets from the exchange.

Again, this is showing off the update workflow. So here, the architect might be adding some windows, doors, different pieces to that model. That model gets republished. And in Inventor, the update can be pushed back into Inventor.

It was a little fast. So I'll just show that right there. So that's one of our first integrations that we released earlier this year.

And I wanted to show basically the same integration. But here, instead of going to Autodesk Docs to create the exchange, the architect can do it right within Revit itself. So it doesn't have to go to that extra step of loading the web page to Autodesk Docs and creating the exchange. So I'm going to play this video.

This is using our Revit connector, which is also in public beta. And we're going to be releasing in-app connectors for many of our integrations because many users just want to work within their app that they're comfortable with. So here, you'll see that that exchange gets created.

It gets updated in Docs, which is where the exchange is stored. And then the Inventor user in the bottom right can check if there are any updates and pull that data in. So see, it was just a quick jump geometry manipulation. And here, we'll go ahead and manipulate that geometry again. And you'll see that those updates are pushed pretty quickly into the exchange databases on Autodesk Docs and then pulled into Inventor as well.

So what's great from this is we're going to be releasing this to developers as well. So you'll actually be able to build on our out-of-the-box connectors in Revit, in Inventor, and some of our other Autodesk tools. So boom, there's that second update that was pushed.

Now I want to show one of my favorite integrations, which is our integration with Power Automate. So this is an exchange integration with a Microsoft, a non-Autodesk tool. Power Automate is a workflow automation tool.

So you can set up a trigger with a set of actions that basically automates anything from simple notification workflows-- so if Farzad were to send me an email and I want to send an automated message to him saying, hey, I got your email, I can set that up with Power Automate. You can also do more complex workflows like writing data to an Excel table or to a Power BI dashboard. So we hear a lot of customers who want to be able to automate some of their business processes with some of the design data that they're getting from Revit and other tools.

So again, very similar workflow here. We're designating the 3D views from which we want to create an exchange. We upload that Revit model to the cloud.

And then we can create an exchange from any one of these 3D views. Again, similar process here up into this point. Here, we might want to just share the windows and wall data to notify someone to create a dashboard or to write a cost report in Excel.

Once that exchange has been created, the user then can go to Power Automate, use the set of templates that we have published on the Power Automate platform, and create, again, these automated workflows. So here in this workflow, we're setting up a trigger that says when an exchange is updated, go ahead and post the message to my colleagues in Microsoft Teams that we've made an update to an exchange.

You can also, like I mentioned, create reports in Excel, which is what we're doing here, and we're writing different elements and different parameters of those elements to that Excel sheet. Now, once we run that automated workflow we see we get a notification there through Power Automate. And then our Excel table also gets populated in real time with data from that exchange.

You can also create dashboards with Power Automate. So here, we're seeing reviewing just data of the exchange, which includes just window and wall data. So this can be useful for reporting project status and really specified information about a building model.

Now let's go to one of our other integrations that's been asked for a lot from our conceptual design customers, architectural customers, which is Revit and Rhino. Again, a very similar workflow here-- pretty streamlined. Open up the Revit connector, select the data that you want to share. Create an exchange-- again, that's stored in Autodesk Docs.

And then you can move over to Rhino and actually use our data exchange connector in Rhino. Here we're showing something that's early phases. But by the time AU hits, this will be in public beta. And you can pull that data exchange that we just shared in Rhino.

You can bring in that geometry data as well as some of the parameter category family data as well. Now, we've got a unique ask from customers to provide some mapping-- mapping categories, families to layers in Rhino. So here we're just showing you have that sort of flexibility to pull in some of that parameter data from Revit.

Now we've heard a lot, hey, you know what? Most of my workflows, I actually want to write data from Rhino to Revit, or I want to write data from Rhino to Inventor or from Inventor to Rhino. We hear you. We hear that that is a super important part of your workflows and something you're looking forward to. We're going to be delivering those right capabilities over the next year.

Now, we've had a few early customers start building their own connectors, and you can actually do that today by going, like Farzad mentioned, to boards.autodesk.com, taking a look at the Data Exchange APIs. We'll be publishing a connector kit on there in the next couple of months, where you can get started as a developer building integrations into data exchanges. So you can start, again, into your custom app, pull in just the data you need that's created from Inventor, Revit, SolidWorks, a host of different applications.

Now, we have a couple integrations customers have built, integrations into PowerPoint, integrations into Tekla. These are all proof of concepts but just showing how quickly you can kind of get up to speed with using data exchanges. We also have customers who have taken the out-of-the-box workflow, that Revit-Inventor integration I talked about.

They've been using the API to layer on even more information that they weren't able to get from just the out-of-the-box workflow, which really focuses on geometry in this case. So there's a lot of different options. We're curious to hear from you which connectors you want to see Autodesk build out, which connectors you want to partner with or contribute to.

We really want to build this ecosystem out with you. Now just a couple of things on our road map for when looking at some of the end user applications of data exchanges. We're definitely building out more connectors.

We're working with partners. We'd love to work with you on seeing how we can build out this ecosystem of connected applications all the way from design to construction. Right now, we're focused on kind of the design disciplines but definitely want to get into some of those business and construction applications.

We're going to be working on enhanced filtering and search. So on the end user side, once you've got your exchange, how do I filter the data down even further, and how do I search on that data really efficiently as well? One thing we've heard from early customers as well is how can I create templates around exchanges.

What if I want to-- I've got a consistent kind of model or information I want to share with customers. How do I repetitively create exchanges just based off an out-of-the-box template that I can create? So we're going to be working on that too.

Now, for developers, our exchange read APIs are in public beta. We're going to be delivering a GraphQL interface, something that Farzad mentioned earlier, in the coming months as well as write APIs that are going to be coming in public beta. That connector kit help you get started. It's got some UI frameworks you can use, some different SDK components that will make it easy for you to integrate data exchanges with your own application.

And here's just a quick glance. I'm not going to spend a ton of time on this slide, but here's a quick glance of what connectors we have out-- Revit connector, Rhino connector in public beta. The Inventor and Power Automate integrations as well are out there. We're going to be focusing on building out Dynamo Grasshopper.

So it's for some of you low coders out there who are doing some really interesting things with filtering and on Revit models, for instance, or on Rhino models. We're going to be providing connectors for those applications as well as getting into some of the infrastructure and structural apps that you see out there. Some of them you'll see here are non-Autodesk solutions. Many customers are using Autodesk and non-Autodesk solutions, and we certainly don't want to be the barrier to using those other tools in your workflows.

Now just-- I'll end on this slide. It's just how you can get started. So Farzad mentioned go visit our website.

Go try out data exchanges today. As end users, you can go do that. All you need to actually create an exchange is a Docs license.

You need Revit 2023 to author the actual exchanges themselves. So that is a caveat there. But it's out in the wild.

Go out and try it out today. Reach out to us at those email addresses you see at the bottom for any questions you might have. And if you're a developer, you can go and visit forge.autodesk.com and try out those data exchange and Fusion Data APIs.

Another special ask we have-- if you are really interested in trying out early technology, early products we're putting out there, you want to participate in research, reach out at vanguard-program@autodesk.com. Our Vanguard program is where we invite customers, developers, as well as just end users to participate in early research, early adoption, early access. We have a few, I think 20 partners there at this point in time.

They've been critical to providing us with early feedback on rolling out these APIs and this functionality. So we want to hear from you. Feel free to reach out to us if you want to participate.

And of course, we'll be having, as we usually do, accelerators for everyone from highly skilled developers to folks who are just starting to code. You're coming from industry. You're maybe an engineer, a fabricator, and you want to learn to code.

Those accelerators are for you as well. So we'll be hosting those in the future. And reach out to us if you have trouble finding when the next accelerator is taking place and where.

So I think we'll end it right there. Thank you so much for listening. Again, don't hesitate to reach out, and have a great AU.