説明
主な学習内容
- Learn about the value of digital twins
- Learn about how fellow customers are using Forge to build digital twins
- Learn how to get started creating your own digital twins
- .
スピーカー
- Jessica Di ZioJessy is the Senior Product Manager responsible for the Forge Viewer and its advanced components. She is an entrepreneur by heart, an engineer by training, and a PM by practice which brings a mix of traits to solve the toughest customer problems. The new challenge she has now is to provide tools for innovators to accelerate digital transformation to sustainably solve design, make and operation challenges.
- Michael BealeMichael Beale is a Senior Developer Consultant since July 2017 for the Autodesk Developer Network and Forge Development Partner Program. Before joining the Forge Platform Team, Michael worked on Autodesk Homestyler, Cloud Rendering, Stereo Panorama Service (pano.autodesk.com) and A360 Interactive Cloud Renderer before working on the ‘Forge Viewer’ and Viewing APIs. Twitter: @micbeale Blog: https://aps.autodesk.com/author/michael-beale
- Shawn WeeklyOver 17 years of progressive experience in system architecture, software development, and innovation/thought leadership for national utility, telecommunications, health care, marketing and engineering companies. Over a decade of current experience in the electrical utility space doing software/system architecture, systems design, and volunteer consulting across a broad range of design technologies and related infrastructure. Deep believer in the idea that with good BIM data and high quality models reflecting that data, we can change the way we do our work for the better, regardless of our current industry. Strong background in the Autodesk design tool stack, Microsoft Technologies, Service Oriented Architectures, .NET and Internet/Intranet/Windows development.
- MGMani GolparvarDr. Mani Golparvar is COO and co-founder of Reconstruct Inc., a venture capital funded spinoff company from University of Illinois at Urbana-Champaign (UIUC). He is also Associate Professor of Civil Engineering and of Computer Science and Faculty Entrepreneurial Fellow at UIUC. He received his PhD degree in Civil Eng and MS degree in Computer Science from UIUC in 2010 and MASc in Civil Eng from University of British Columbia in 2006. His work on computer vision data analytics and BIM has been the recipient of several awards including 2018 Walter L. Huber Award for Innovation in Civil Eng from ASCE, 2017 ENR Young Professional and national Top 20 under 40 Awards, 2016 ASCE Dan Halpin Award for Scholarship in Construction, 2013 ASCE James Croes Medal for innovation in Civil Eng, 2013 and 2010 FIATECH CETI awards. Dr. Golparvar's work with Reconstruct has also been recognized with 2017 World Economic Forum Innovation Award and 2015 Innovation Award from Turner Construction.
- Cindy BaldwinCindy Baldwin is the President of VDCO Tech and VDCO Data and has 30 years of diverse project experience. She has a proven ability to lead her teams of BIM, construction professionals, subcontractors and design professionals, while maintaining strong client relationships. Cindy’s key strengths include reducing risk for her clients and increasing efficiency in production through BIM & VDC. Cindy is recognized nationally for her expertise in VDC process and implementation strategies. A frequent speaker at conferences, Cindy is highly regarded for her practical “real-world” approach to BIM implementation and process. She is known for getting into the “BIM Trenches” and working through traditional processes to implement change. Cindy is a certified general contractor, CM-BIM, CM-LEAN, LEED AP, an Instructor for the AGC CM-BIM Classes, Chair of the USIBD Building Data Committee, Chair of buildingSMART USA's Airport Workgroups and on buildingSMART USA's Executive Committee.
Hi, I'm Jessy Di Zio, and I'm Forge Viewers product manager here at Autodesk. I have to ask, have you ever run out of battery or signal on your phone without finding your way back? I have, and it got me thinking, whoa, it wasn't too long ago when we were using paper maps to go from point A to point B.
And it wasn't until Google Maps and MapQuest came to life that totally disrupt the way that we go about places now. We strongly believe that digital twins are the new Google Maps and are disrupting the way that we organize BIM data. We're going to take a deep dive into data with Forge and digital twins.
By the end of this talk, you should have a good understanding of what digital twins are, some of their use cases, and their benefits. We have three amazing customers sharing their stories on how they were able to save money, time, and even life. And then Michael Bell will give us a technical overview to get you all set up to get started with Forge today. So let's get started.
Why digital twins are so popular these days? Before I answer that, let's talk about what digital twins are, are a digital representation of a physical entity. And to better illustrate what they are, let me walk you through an example. Imagine taking a photograph of the building, but a very special photograph, a 3D one. You take one today and then next week and so forth. And this will allow you to predict future by comparing the difference over time.
And when you do that, you can start tracking activity, how things are in the building, compare before and after, and so forth. If I took a photograph at a year ago and compare it to today's, I can see that maybe a plumbing is starting to leak or that mechanical piece is wearing out and it's time to replace. I can run maintenance analysis and prevent future fault.
And just like Google Maps, digital twins are disrupting how we organize and access BIM data, but what do I mean by that? Keep your BIM data organized. They can take lots of different file format, Autodesk or non Autodesk without the need of those applications. Digital twins centralize all of those files into a single place similar to a database allowing them to be unlocked and bring them all together. It brings access from anywhere. You can access Revit file from the browser, and you don't need the Revit application installed on your machine or device to access it.
And I'm thrilled to have three amazing stories to be shared today by our customers of their journey with Forge from the problems that they were trying to solve to the implementation to the outcome and success. Our first customer is Reconstruct where Mani and Cindy will walk us through how they analyze things over time with past and present to predict the future with drawn videos and laser scan on a project in Florida.
MANI GOLPARVAR: In the next few minutes, Cindy and I will be speaking about Reconstruct and the construction story, construction around Broward County Convention Center, basically covering three or four topics, the construction problem, an overview of what Reconstruct is and how this Visual Command Center operates, and some lessons learned both on the technology side of Reconstruct and also the project itself. Cindy?
CINDY BALDWIN: Thank you so much. It's the Broward County Convention Center, and this is located in Fort Lauderdale, Florida, five miles from our International Airport about a mile from the beach right there at the port. Broward County partnered with Matthew [INAUDIBLE] for the project as the developer.
It's a design build project. Stantech is the architect of record. Balfour Beatty is the contractor. And the Whites Company is the owner's representative for the project. My company is VDCO Tech, and we're part of the owner's team. We are providing construction administration services, VDC BIM oversight, [INAUDIBLE] data integration, and schedule review services.
This project had to open in 2021. It had a hard deadline because of the boat show. It's a very, very large project. Schedule management and risk management were going to be our largest challenge. After researching the marketplace, Reconstruct was the software that we ultimately selected because it provided not only photo documentation and schedule integration, but it ultimately visualized our progress modeling through the schedule.
So we looked at the problem by each participant, and we wanted to address their pain points. For example, with an owner, we wanted to be able to review major milestones and look at the big picture. To the superintendent, to the GC, we wanted to make sure that we were ensuring compliance with the schedule while overlaying critical path with site progress in the models. For the project schedulers, we needed visualization plan versus actual. And for our trade contractors, we wanted to visualize teamwork. We wanted them to be able to see schedule dependency.
MANI GOLPARVAR: At Reconstruct, we tapped into images, videos, and the design data, and bring all of them into the same integrated environment. You see what work is being planned to be executed for the site, see how much work has been done in the form that is measurable, and then you can contrast it to what is there versus what should be there as the basis of progress tracking as a basis of quality control and proactively identify risk for delay.
For Reconstruct, we bringing the design data in various file formats, and we map it against the schedule of the project. That 4D BIM is overlaid against every type of image, cell phone image and 360 camera with using drone. Now you have a measurable, quantifiable design file that is overlaid on a measurable image. All of that is being visualized in the context of the schedule. Various work packages, tasks, and locations can be color coded based on risk for delay or the actual progress.
There's an ability to form a time machine over project. There are three specific use cases. One, for the past, it gives you an opportunity to create historical record. When you overlay the design on top of reality, you can use that for quality control, checking the spacing between the rebars, conduit placement, [INAUDIBLE] placement. We can use it for progress tracking to plan the work or replan and also identify elements of risk in your project and use that to proactively tap off delays before they happen at the site.
When you bring in images and videos from 360 cameras, drones, and cell phones, Reconstruct has a computer vision engine performs photogrammetry, transforms your images and videos into models and this 3D View experience that is measurable. The second piece has to do with the integration that we have with Forge which gives us 60 plus models that can be imported into Reconstruct. That federated model is brought into alignment with reality data, for quality control, for progress tracking, and payment application.
And the last piece has to do with integration with the schedule. In Reconstruct, we have the ability to tightly integrate with Oracle Primavera Cloud in a bidirectional form. We can also bring in your desktop PC files, Microsoft Project. You can bring Excel. The system is all web based so you can see what work has been planned at any position in time. You see elements that are being color-coded based on work expected to be done in contrast against the reality. Cindy?
CINDY BALDWIN: So for any Back to the Future fans, this is the DeLorean, right? Reconstruct to us was proactive risk management. We wanted to visualize that risk, what was going on there and what should be going there at any specific time in our schedules. This chapter is January 18 of this year. We wanted to get a close-up view of an area that we had a concern. We were tracking some structural steel installations.
We were able to simply click on the image, get closer views to be able to look at areas that had potential critical delay. It all revolved around setting that first long-span truss, and we were closely monitoring the structural steel installation in the sequencing of that against the project schedule.
The efficiency we ultimately realized as a [INAUDIBLE] team and then allowed my field inspectors and Reconstruct coordinators to dive deeper into captures. We could use the BIM overlay tool as you just saw, and we can ensure compliance, but we could also check dimensions, and we could look at any kind of discrepancies and deviations that were actually happening that were ultimately going to result in better, more accurate as-built documentation, which was key.
On these future slides, that same exact area on the scene capture date. They were able to easily see what is there on site to what should be there based on trade contractors' installation models, and then we compared it against the contractors' likely schedule updates. By overlaying the BIMs and with this color coding, we can see what activities should have started and been completed based ultimately on the schedule. The green items that were showing up in the models are actually activities that should have been done based on this capture date. And you can clearly see from the video that we're behind schedule. We can now communicate to all stakeholders the areas of exact concern.
Moving forward, my scheduler never had to go to the actual site. This is a view of the same area but from a different floor. The scheduler and project manager simply reviewed the weekly photo captures as part of our scene services. We then inputted the percent complete based on Reconstruct's field captures and scheduled activities. Doing this work remotely then saved travel time, provided us a safe work environment especially with COVID and site closures.
I spoke earlier about the concerns we had with those trusses. If we look back on those capture dates, we're able to see where those dependencies are in the schedules, comparing it directly to the models in the same visual overlay. You're going to see where it is on the bottom. And we're able to then say, we're no longer working in that vacuum, and we can toggle back and forth between the dates of the schedule versus the dates of the actual. We're being more proactive, and we can visualize that project risk based on our site conditions in the captures.
If we went off the typical schedule revealed, my field inspector never saw eye to eye or understood each other's concerns. We were literally working a month ahead of each other. The time we were able to put the schedule in the field inspector into Reconstruct, our communication and collaboration was instantly improved.
We still do our monthly reviews based on the schedule day to day. However, Reconstruct is what we're using at all executive meetings to show actual date and actual field delays to see if it really should be something that we're concerned about or if the contractor had already made an adjustment as we moved forward.
MANI GOLPARVAR: Thanks, Cindy. So as Cindy pointed out, these three specific problems, that meant we had to deal with a lot of data. We decided some time ago to build our system on top of Forge, specifically, two capabilities. One is the ability to tap into Auto Derivative API to make sure that the device works, Revit is being extended in a life of the construct as a basis of progress monitoring and quality control.
The other aspect of it was the viewer because these are very large models. Things that we demonstrated from a computer vision perspective, automated generation models, those are completely unique to Reconstruct, but we are able to bring that together with Forge. We also had to worry about efficient way of rendering point cloud data, how we can automatically align them against the design both in 2D and 3D. Cindy?
CINDY BALDWIN: As contractors and trade contractors, it made us transparent. It ultimately helped us provide better value to our clients. And as an entire project team, it helped us create efficiencies in our workflow. Bottom line for me, Forge and Reconstruct provided that transparency through that digital twin. It enables us to provide those better services. And for that, we're truly grateful.
MANI GOLPARVAR: Thank you, Cindy. And by that, digital twins for us means images and videos that are being automatically transformed into reality models and are being tied against the design for actively forecast of future and use that as a basis of risk management. Thank you for the opportunity.
JESSY DI ZIO: Thanks, Mani and Cindy. Your solution shows how technology is helping build better, faster, and safer. Thanks again for sharing. Next, we have Anand and Adeel from Accenture with their connected construction app, how they organize massive amount of data from multiple systems and bring them all together to help a major oil and gas customer.
ANAND DESAI: Hello and welcome. My name is Anand Desai. I'm from Accenture Capital Projects Two based out of Houston. I'm also the product owner for our Connected Construction platform. And today, I have with me Adeel Ali also from Accenture Capital Projects Group Houston office. He is our Connected Construction data visualization lead.
Connected Construction is a platform that helps organizations realize the full potential of their people, equipment, and materials by connecting them in real time to improve product safety, efficiency, and most importantly, to make their project delivery more predictable. This POC, what we did was using Connected Construction for one of our oil majors out of Houston.
The situation was the client was getting into a multibillion dollar, multi-year offshore platform construction where they had multiple stakeholders, multiple EPCs, a couple of fabrication yards, and then engineering teams spread across US, India, and Europe. So they had a lot of information which are static and inconsistent, multiple source of truth, project reporting was delayed, and most importantly, a lack of accurate timeline sequencing of their work packages.
And what that was doing was they had very inaccurate demand planning of their resources on the site. Teams were reactive, meaning they were looking at lagging indicators which were coming from the construction site. So they were always behind. They were seeing a lot of change orders, rework, and wastage on site, which were all resulting in negative impact on the productivity and progress. And most importantly, what it was doing was making the construction site unsafe for their people to work.
We did a 10-weeks POC with our client using Autodesk Forge and Connected Construction to create four different modules. The first one is model visualization. Second is construction progress. Third is path of construction. And fourth is system completion. I'll go into each of these briefly and then let Adeel do the complete demo of this particular POC.
The first module that we built was a model visualization. So here, we got the 3D model from the EPC vendors on the project, and then we integrated the 3D model to our owner's project information management system. So users could click on any single component on the model and then get all the relevant information that was in the background.
The second one was construction status. Here, the integration was between the 3D model and then the P6 schedule. So whenever the P6 schedule was updated, you could see what was happening in near real time, what tasks were completed, what were on time, and what were delayed.
The third module that we built was path of construction. This was extremely useful for them during constructor building reviews. So again, the integration here was between the 3D model and the P6 schedule, but here we were trying to visualize from start to finish every single activity, every single work package in terms of when it was being done. What it did was it helped our clients in terms of sequencing, understanding procurement requirements if there were any clash, and it helped them manage multiple contractors on site with their people, material, and equipment.
The fourth one is systems completion. What we did was we integrated the 3D model with the system's completion module for the project. We were able to visualize every single system that was part of the project and the status of each of these systems. The client had around 240 subsystems which were tied up to 18 systems. So they could visualize what's happening with each of them. They could pinpoint which work package was slowing them down, which was on time. If there were any resource constraints, they could visualize all those things in near time and collaborate with it. Now Adeel will go through a demonstration of the POC that we did with this client.
ADEEL ALI: Hi, everyone. I'm going to give you a demonstration of the four features that Anand just covered. What we're looking at here right now is a model of the top side. The model was originally in the Navisworks format. It's been converted to the Forge Viewers as we have formatted using the model to Rivet API.
The model and associated data feature is basically your standard 3D viewing functionality. You can orbit around the model. You can zoom in as well as make use of the Forge toolbar here. So we've got things like the hierarchy, change the camera angle, take measurements, et cetera.
The only thing that we've implemented in here that's different is a Properties window. It's a click event so once you click a component, it'll pull data from the asset register, give you the tag ID, system information, as well as some additional properties. If there's any documents or other fact files associated, it'll pull up a link for those items as well.
The next feature is the construction progress. So as I switch over to the construction progress, you'll notice that the model does not load again. We have the HTML and CSS implementation in here where the only thing that's changing is the overlays, and the model doesn't load again. Construction progress itself is a view into what the status of the four package installation is is for the current week. But we could also cycle back to previous weeks as well.
So if I need to look at something, for example, that they've highlighted, I can double click it, and it'll zoom in and allow me to look around the specific component and see if there's anything else that's delayed around it. I also have the ability to hide some of the coloring. So if I'm not interested in the particular status, I can turn that off.
The next feature that we have is path of construction. Path of construction is a view into the sequence of installation of the work packages. So if I'm interested in a particular window, I can pick the dates from the timeline view. For demonstration purposes, I'm going to pick a particular month here so that we can quickly get through and view the path of construction.
So once I click View, you'll see a list of work packages by week that are scheduled to be worked on. Across the bottom is a similar view. I have the number of weeks, the number of work packages. I can pick a specific week if I'm interested as well as go and toggle between each sequence or work packaging here as well.
What I'm going to do instead here is press the Play button, and you will see an animation of the work package that's being worked on. So anything that's being worked on in the current week will get highlighted in green. By the end of that week, let's say that that equipment is fully installed. It'll shade to blue, and anything else that's being worked on in that week will turn to green.
I'm going to close it here just for a moment so that you can see that you can still navigate around the model. You can still zoom in, zoom out, take measurements. One of the benefits of path of construction is being able to see the sequence of installation and identify any clashes. So that's where measurement tool comes in handy.
The other thing that you'll notice is that the rest of the model aside from the equipment that's being worked on during this one-month window, everything else is transparent. We're using the Viewer's ghosting feature to do this. So anything outside of the filtered window is ghosted. Anything that's being worked on will be shaded in the color overlay that we designed.
The last feature that we have in here is the system completion. System completion is a view into, post construction, how systems are coming online and with the statuses. We track that status by looking at the checklists and punch lists that are being worked on. So for system 210, for example, I have a breakdown by the system and subsystems. And within subsystems, I can look at the checklists and the punch list and what their statuses are. And that's what's being reported on here.
So those are the key features that we have for digital progress tracking within Connected Construction. Overall, it was an easy mutation and easy integration with the Forge APIs. That's because Connected Construction itself is a cloud native tool, and Forge makes use of modern web services so it's been relatively easy to integrate.
One challenge that we did have initially was we were using the SVF format for this model, and a lot of our clients and their process are industry heavy, and they have very dense models. So they have a lot of piping and a lot of vessels, things where geometry may be similar, and the Forge team was able to get us onto the beta program for the SVF2 format. And getting on to the SVF2 format was pretty easy. There wasn't a lot of rework required. But after we switched over to SVF2, our performance was greatly improved.
ANAND DESAI: Outcome of this POC engagement was that our client had single source of truth, real time data for accurate and faster decision making, focus on leading indicators that were affecting their project cost and schedule, improved asset planning and utilization, a clear path of construction so they know if there was any clashes, if there were any issues with their procurement routes, and most importantly, increased confidence on their project schedule.
ADEEL ALI: So that concludes our demonstration into the Connected Construction's use of the Forge platform and its various APIs, extensions, and methods. Should you need to reach out to the Accenture Connected Construction team, our contact info is in the handouts. Thank you.
JESSY DI ZIO: Thanks, Anand. Thanks, Adeel. This was incredible to see how you guys helped this customer get really organized and have easy access to the data. And last but not least, Shawn from Southern Company will tell us how digital twin is helping keep his team safe from being shocked by electricity. Take it away, Shawn.
SHAWN WEEKLY: Hi, everyone. My name's Shawn Weekly, and I work for Southern Company in Birmingham, Alabama. And we are using Autodesk Forge to hopefully prevent things like this from occurring. This is what happens when you open up a 500 Kv switch that has a bad circuit breaker. Whew, indeed, right?
So how do we prevent stuff like that? And in particular, how do we predict equipment failure due to operational or mechanical issues? We want to be able to extend the life of our equipment and overall really just be able to reduce power outages because we don't want people to be without power.
We use Autodesk Forge in a program that's called Condition Based Maintenance. And for us, that means we take our 3D models. We put those in a browser. We capture and use our IoT sensor data, and we link in all the supplemental information that we have available to us and put all of that in front of our users so that they can make decisions and do proactive maintenance.
So we start off with models in a browser, and we use Autodesk Inventor. We like to use Autodesk Inventor because we get really detailed models. We can count sheds on insulators. We can get down to the bolts, nuts, and washers. But we also, because we use Autodesk Forge, we can do markups and measurements, and we can collaborate with our users. And that's important. That's not something that we've had available to us in the past with paper drawing.
We put IoT sensors on all of our major equipment. In the picture here, you see a transformer that's got a couple of sensors on it. And we want to read the major equipment's vital information. We want to also be able to visualize that data, put it in front of our users, and let them do near real-time analysis and proactive maintenance decisions so we can proactively decide to go ahead and do some work on a piece of equipment that's not behaving or is trending up and down on a vital number.
And then we take all of the supplemental information that we get. We use Autodesk Vault on our back end. And we capture property information from our project's upstream systems. We put that in properties in Vault, and then we send that back to Forge. And we put all that stuff on our models, and we have that available to our users. And then this next video that I'm going to show you, I'm going to show you how we bring all that stuff together.
This is a version of our power delivery digital twin app that I have running in my local development environment. I'm going to click on my station's link here and show you I've got four models loaded into the Forge OSS buckets. Below the web page, you'll see that I'm logging some information just to the display window so that I know where I'm at in the application and to keep myself in track. That
Navigating around in my model, the level of detail is really good. The performance is really good. We're converting our Inventor models to the Forge SVF file format with no loss of detail that I can tell. Performance in the browser is great. I can count bolts. I can count sheds on the insulators. I have really, really good outstanding picture of my [INAUDIBLE].
This particular substation is a little bit bigger than the other three. This is a transmission substation versus the other three. They are distribution subs. Really, the only difference is the level of detail and the amount of voltage that each one can handle. But I use this as a model because we can see just a huge amount of detail still. There's just a lot of really good stuff in our models that come across.
At our Airport Road station here, as I Zoom in, I want to just peek at my regulators and my breakers. But what's important here is I'm going to show you my transformer and click on it. And I want to bring up the properties that we pull into our model from Inventor with Autodesk Forge and Autodesk Vault on our back end where we integrate with our systems record, and we pull lots of information into the model for our users.
So things like our emergency address and lat and long for GIS purposes. We have voltage information. We get unique equipment IDs, the equipment descriptions, and a lot of project-level details that our customers want to see so that they can keep track of all the projects that are ongoing within the station at any given time. And that just becomes important for them to make maintenance decisions in particular so they know where to go and to do.
We get all of the same Forge-level goodness that we get that the Revit guys get in AEC with our Inventor models. So when they're converted, we still get markups and measurements. We can get accurate measurements here on the underground bus work. We can measure the height of our poles and check clearances. And when we enable the collaborative features, we can export these out as issues even linking it back to BIM 360.
With the Enable Sensors widget, when I click here, I've got a breakpoint set in my JavaScript code, and I'm just showing you some of the bits. But the more important aspect here is that Forge is written with modern languages in mind. You're not going to feel overwhelmed or have to go find somebody who knows some kind of weird programming language. This is all modern web stuff. And you can get development talent to work on here. We pull our data in from SQL Server on Azure, but that's just what we're used to. That's what our development standards are based on.
This particular big graph here pulling time-series data from SQL Server from Azure. We're measuring the temperature in this transformer. And I can see peaks and valleys that our customers can make decisions based on whether they see that as a problem or not.
So as I hover over the sprites here, the Data Vis API for Forge gives me hooks back into the API, and I can set more break points, and this is in my .NET code just to show you that, once again, modern language support. .NET core is very well supported. I'm more of a .NET guy versus JavaScript so I'm comfortable in this environment, but regardless, whatever developers you have, whatever language they're comfortable in, you can use this technology. As I click on these sprites, I can bring up more sensor data here, and this sensor data is fed into the system with three second loop to show you that we can indeed show near real-time data to get different values coming back so we give this to our users for them to use.
One of the last features I want to show here is some supplemental drawings that I pulled into the system. This is a 2D AutoCAD electrical drawing that shows a three line connection plan for this substation. You can see the level of detail's still great viewer in this 2D asset just as well as it does 3D. And we can turn on the properties and the layers if we so choose.
And we're not just limited to black and white drawings. We can pull in color drawings as well. This is from our physical design shop, and they do their stuff in color. The accuracy of the drawings is great. So if this drawing is drawn to scale, you can get markups and measurements just as well, and you can continue to collaborate with your customers, save these assets out, give it to them so they can make business decisions. Thanks.
JESSY DI ZIO: Thanks, Shawn. I'm glad you guys were able to leverage digital twins to keep your team safe. Now hopefully, you are excited and wondering how can I get started today. Well, Michael is going to give us a brief one to one on how to build your own flavor of digital twin with Forge. Michael, over to you.
MICHAEL BEALE: Thanks, Jessy. I'm Michael Beale, developer advocate for Autodesk Forge team. Today, I'm going to get you started with building your own digital twin. We're going to build it in three steps. First, we'll organize your data. Then, we'll make it easily accessible. And lastly, we'll customize it. Let's start with organize it.
So what do I mean by organize it? Well, I mean we're going to bring all of your CAD files together, convert 70 plus different file formats. I'm going to bring all of those files together into one place. And we're going to unlock that data by converting them into a normalized format using the Forge Model Derivative cloud service.
Once we've converted those files into a normalized format, we can tap into BIM geometry and properties metadata, and that's going to make it useful for combining different bits of data with other systems like a cost system for quantity take-off or measuring, say, door clearance for on-site equipment logistics. Model Derivative is a Forge cloud service used by ACC and BIM 360.
Second step, how can my team access this data? The Forge Viewer is a module that makes it easy to view large models in a browser from anywhere. It's a JavaScript module based on Open Source 3JS so it's easy to customize. So if you're familiar with 3JS, you'll be familiar with the Forge Viewer.
Now I'm going to cover the basics of the Forge Viewer. I've got a Chrome browser open to a web page. Type in dasher360.com, and I'm going to click on the Demo button. So this is the classic digital twin project, Dasher 360. You can see that it's using a stock version of Forge Viewer. I'm going to use this to demonstrate the Viewer API basics.
First, let's cover isolation. I've got my model. I can spin it around, and I can isolate individual things. For example, I just want to see the first floor. Click on the user interface, scroll down to level one, and I can isolate level one. But I can also isolate individual objects. So let's say I wanted to isolate this desk and chair. I right click, and I select Isolate from the UI, and now I can just see the chair and desk all by itself.
I can also open up the model browser. The model browser has a tree hierarchy of all of the objects in the scene. I can select on individual objects and turn their visibility on and off. And I can do that because Forge Viewer has a render match system, a progressive renderer, a BVH, and a streaming mesh protocol that make isolating individual pieces and parts easy to do. Doing that with stock 3JS is going to be a lot of work, and so Forge Viewer is going to save you a lot of time.
OK, so how do I programmatically isolate things? Let's open up the Chrome Debug Console. I'll bring up this little window here. And let's find an object to isolate. I'm going to type in the command Get Selection. So now that I've got an ID, I type in the command isolate with the number, and you can see that's now isolated. Now it's a little bit far away so I'm going to use another command called fit to view, and that's going to zoom the object in and fill the screen.
So let's look at the coloring features of Forge Viewer. I'm going to do that with a command called set theme color. So first, let's select a whole lot of walls, and you can see I've got an array of IDs. OK, let's type that in, and now I'm going to set theme color. I hit Return. Great. Now, I can see all of those walls have turned red, and I can use this for, say, dashboarding.
All right, let's refresh the browser now, and let's cover some more advanced things. So I can turn on Show Sprites, and those sprites are actually sensors. I can turn on the occlusion, turn on heat maps. I can colorize things based on things like temperature or presence. You can see here people are moving around in the building.
I do this using an extension called data vis. So for template source code of this digital twin, go to the Forge documentation, click on Documentation, scroll down to Data Visualization, select Developer Guide, and then on the left, click on Quick Start. You can also click on Examples. There's examples for sprites, for animations, and also for different sorts of heat maps. To get help from the Autodesk team, send an email to forge.help@autodesk.com. Now back to you, Jessy.
JESSY DI ZIO: Thanks, Michael. That was a great overview. And with that, I would like to point you to where you can get started today. So open any tab and go to forge.autodesk.com to start a free trial. Thanks so much for joining, and don't forget that Forge at AU has more interesting content so check them out. Don't miss a roadmap workthrough, some of their cool lightning talks, and now we have a Forge documentary, and of course, the Hackathon Showcase. Thank you so much.