Description
Principaux enseignements
- Explain the Manufacturing Data Model (MDM) and what you do with it.
- Building Automation using APS Services viz. Authentication, Data Management, Model Derivatives, Viewer, Design Automation etc.
- Fetch physical material properties fetch for analysis (viz. Sustainability) and writeback results to design using MDM.
Intervenant
- Rajesh BhartiyaRajesh Bhartiya has been in the engineering software industry for 25+ years now. During this period he has developed numerous software applications in the field of CAD/CAM and engineering graphics. At IIT, Mumbai, where he completed his Masters in Aerospace, he developed an aircraft cockpit simulator, which fueled his interests in 3D modeling and graphics. After completing his M.Tech he worked as a 3D graphics and geometry programmer at Geometric Software Solutions Limited (GSSL) and with Parasolid development team in Cambridge, UK. He went to the US in 2001 and joined Tech Soft 3D (TS3D). At TS3D, he not only developed new products but was instrumental in setting up internal processes, which benefited the company's engineering efficiency. Rajesh proved to be a very quick learner and filled many key roles ranging from core developer, new product designer, demo creator, and on-site consultant to many companies. In 2005 Rajesh saw companies needing to develop advanced engineering applications but at the same time lacking the skills or resources to develop them quickly. With a vision to leverage his broad spectrum of engineering software skills to help such companies quickly find and adopt a solution, he founded ProtoTech Solutions. Based in Pune, India, ProtoTech already has a rapidly growing team of engineers working on various new technologies. He is now happily settled in Pune, India with his wife and two kids. Rajesh enjoys his time most when he is designing and developing software. He is a voracious reader and an adventure nut. He has summited Kilimanjaro, cycled 600 kms in the Andes to Machu-Picchu and ran Antarctica and Everest full marathons. Read further about me here - https://prototechsolutions.com/rajesh_bhartiya/
RAJESH BHARTIYA: Hello, everyone. Thank you for joining me today. I am Rajesh Bhartiya, founder and CEO of ProtoTech Solutions. I am passionate about software development-- each and every stage of its life cycle, from designing to developing to maintaining and enhancing. I studied mechanical, aerospace, and computer science engineering, and I have spent over 2 and 1/2 decades leading many engineering technology products.
Along with the fun building with APS-APIs, I also enjoy another APS-- Adrenaline Pumping Sports. I have summitted Mount Kilimanjaro. Cycled 600 kilometers in the Andes to Machu Picchu. I have also ran a few extreme marathons, like one in the Antarctica and another which starts from Everest base camp.
This and the first APS-- the Autodesk Platform Services-- both the topics excite me very much, and I'm always looking forward to bond over them any time. My company ProtoTech specializes in custom software product development. We are experts in engineering technology. We can build, integrate, automate, customize applications in this space.
Founded in 2005, we have over 100 experts. And in last two decades we have contributed to 200-plus projects spanning various domains such as architecture, aviation, agriculture, additive manufacturing, and so on.
ProtoTech and I personally-- we both have decades long connection with Autodesk. My first association when I signed up for AutoCAD course back in 1995. And over the next few years, I continued to develop apps which used ObjectARX, [INAUDIBLE], DWF Toolkit, and so on.
Within a few months of founding ProtoTech, we became ADN members. And very soon we were top developers. We were one of the first companies to launch on the Autodesk App Store, and we have 100-plus plugins on the Autodesk marketplace. Our plugins have seen 100,000-plus downloads to date.
Two decades later, we still work very closely with various teams at Autodesk and their customers. We are APS-approved systems integrators and certified partners in manufacturing space and construction industry space.
Well, before I proceed further, I have a disclaimer. My role in this project has been on the business side of things. All this awesome tech work has been done by my smart team at ProtoTech, and the entire credit goes to them. No doubt I'm proud to represent them here at AU. At the same time, I would like to apologize for any errors that I may end up making here.
OK, what are our objectives today? Well, APS has some fantastic tools in its [INAUDIBLE], and I will introduce these tools too, and talk about how to stitch them together to build what you want. I'll particularly focus on manufacturing data model. We will find out what it is and ways to fetch material properties. We will also find out how to write back your own data, like analysis results or et cetera, back to the design so that it persists and it can be shared.
Now, my class title has sustainability included, but it doesn't show up on the objectives. Well, let's talk about sustainability, too. We all know that the design phase is always a key phase in product's life cycle. And according to the European Commission, it is estimated that over 80% of all product-related environmental impacts are determined during the design phase of a product.
Now traditionally, the designers are used to performing various analysis-- stress analysis, thermal analysis, vibration-- and incorporate the feedback right before the design is finalized. But what about various sustainability parameters? What if the designers knew how much greenhouse gases will be emitted? How much water? How much land will be used? What can and what can't be recycled?
What if the designers can tweak their designs immediately, understand the impact-- the change caused by this tweak? What if the designer can know all this, do all this in real time without leaving their design environment? This can be hugely empowering for them in designing not just for function and aesthetics, but also for sustainability.
So what's stopping us from doing that right now? The problem is that while there are billions of products, the experts are only in thousands. When I say experts, I mean sustainability experts. These are the people who have knowledge and skills in assessing how products affect the environment. They use special methods and tools to do their assessments. They help find ways to make product better for them.
Let me introduce you to LCA-- life cycle assessment. In this method, for any given product or service, every aspect is reviewed. What is taken as a resource from the environment. What is thrown back to the environment. We also assess the environmental and social impact. They are computed over the entire life of the product from cradle to grave. This LCA is standardized, transparent, and scientific methodology, and hence, it's reliable.
Pré is a Netherlands-based company and maker of SimaPro, one of the world's most widely used LCA software. With 12,000 plus users across 80 countries of the world, it is trusted by industry and the academia. Now incorporating LCA in the design phase has many benefits to it. It removes the inefficiencies in the entire process by providing real time feedback on sustainability metrics.
Early intervention helps save a lot of cost in making the products clean. LCA facilitates multiple design iterations, providing multiple product options. It also helps with regulatory compliances. And it demonstrates the commitment of the organization towards environmental responsibility, positioning it as a leader in sustainability design practices and enhancing its reputation in the market. Very important.
Now there is a design tool, there is an LCA, so there is an opportunity. We developed EcoDesigner, the app that brings sustainability analysis to the designers. Here, the design environment is Fusion. And we support Inventor as well. The LCA tool I just mentioned. It's Pré's SimaPro. And we use APS APIs for the integration.
Basically, we fetch the bill of materials data from the Fusion hub using Autodesk Platform Services and then pass it to the SimaPro's LCA services. The results of analysis are fetched back and displayed to the designer in a Fusion environment. The designer can now review, make changes, reiterate, reanalyze, and almost immediately check out the impact.
The analysis results, which are final, are written back to the design. This is the overall how app functions. Now EcoDesigner enables the inception of sustainability parameters before manufacturing. That is in the design phase. But it ends up impacting each and every stage of the product life cycle. You can see that. So all that is exciting. Let's take a ride.
[MUSIC PLAYING]
Awesome, wasn't that? Want to make one like this? I'm going to flash back now and share the journey of how we built it. So to build an application like EcoDesigner APS provides a fantastic set of API tools. We have authentication, data management, model derivative, manufacturing data model, which by the way, will be our focus today. Design automation and [INAUDIBLE] APIs.
I'm going to very briefly tell you about each of these, and give you the context of how we applied it to build the EcoDesigner. As soon as you have an idea to build your application, go sign in to your APS Developer portal, create a new app, and register with APS. Once you do that, you will receive a client ID and client secret.
So APS supports OAuth 2, which helps the program get access to the user data without exposing credentials. Using the client ID and secret that you got earlier, your app makes an HTTP call to an OAuth REST endpoint. A token is returned to your app. In making subsequent HTTP calls to various other APIs on the platform, your app must include this token in the request header. Very simple.
Now coming to EcoDesigner, I know at the very beginning of the user journey, we redirect the user to Autodesk portal to perform this authentication. The user enters the credentials to their APS portal. We use three legged authentication token, which will allow us to access and display a list of hubs, projects, and items stored in the APS user account.
We declare that we will need to read and write the data, and we seek the permission. And once this is out of the way, we are all set to do the magic. We now use the data management API to access the data. There are two ways you can access using data management APS. The first one is simply like a data store. You know, like Dropbox or Google Drive. Here, you are not restricted to CAD file formats. It can be any file, even multi media files, such as photos. Videos.
Second, most importantly, you can utilize the data management API to access A360, BIM 360, and Fusion data and projects stored on the cloud. Since the Fusion files are already on the cloud, no upload is required. But we use data management APIs to support Inventor files.
As you know, Inventor is a desktop application. We have to move the files to the cloud for processing and visualization. To achieve this, we utilized the Object Storage Service-- OSS through the data management API and obtained a URL. We also created placeholders by generating signed URLs for the output files.
Next is the Model Derivative APIs. They help us to translate the designs from one kind of format to another. Model Derivative APIs support the conversion of 70-plus formats to the SVF2 format for rendering. Post-conversion, you can also extract the objects. Their hierarchies. Views. Geometry information. You can query properties such as material density, volume, and so on. You can generate different size thumbnails of design files.
Model Derivative API allows the application to generate data for rendering in the browser and for downstream processes, such as maybe order processing, ERP integration, and so on.
Visualizing analysis results in a browser is a key feature of EcoDesigner, for which we use the APS viewer. Now, this requires translating the design into SVF2 format, which is compatible with APS viewer, and it is accomplished using the Model Derivative APIs. Additionally, we also used Model Derivative APIs to fetch physical material properties for performing the sustainability calculations.
And the hero of our case study today, the Manufacturing Data Model. Manufacturing Data Model API allows developers to read, write, and extend product design and manufacturing data that is stored in the cloud. These capabilities allow various cloud-based workflows, such as navigating through the hubs, projects, folders, files and components, and sharing them. Access bill of materials information and integrating with your ERP and ordering system. Generating STEP files for each components.
Accessing properties such as mass volume material information using GraphQL APIs. And writing back your custom properties-- maybe cost, supplier code, carbon use in our case-- using the same GraphQL APIs. Now we will go in depth on the last two items, our usage of Manufacturing Data Model.
First, what is GraphQL? GraphQL is a set of API technologies developed by Meta as an alternative to REST. It's now released to the public and it is open source. GraphQL can be thought of as a query language for APIs. You request data with queries. You create, update, and delete data with mutations which are very similar to queries.
You can use various GraphQL tools to explore, visualize, and run queries. One such tool is the Manufacturing Data Model Explorer. It is an interactive, browser-based user XP interface that facilitates the exploration and execution of GraphQL queries. This explorer also incorporates an autocomplete functionality so that you can quickly build queries.
As I mentioned earlier, one of the key operations that Manufacturing Data Model supports is querying data information. In the demo, you may have noticed the left panel displaying a list of hubs, projects, and different items. This information is stored in cloud, but Manufacturing Data Model simplifies the process by providing a fast, single endpoint for querying this information.
Additionally, we retrieve properties such as the last modified date. Root component version. Its occurrences. These help us to make sure we access the latest design.
The Manufacturing Data Model in Autodesk Fusion lets you attach custom data to the components. Now, we are talking about how to attach your custom properties. We have already queried it. How to attach them to the components? First, create a property definition collection. It's like a group. We call our group as sustainability parameters in our case.
Now next, we define one or more properties to be part of this group. One property would be, in your case, supplier stock, which then can include things like name, description, supplier code, et cetera, et cetera. One key aspect of the property is property behavior. Property behavior is something that controls how design changes affect the property.
So, for instance, if a design is modified-- and let's say material is changed, obviously the CO2 emissions and other sustainability data becomes outdated. That's why we chose Dynamic at Version as our property behavior. Finally, you link the property collection to a hub using the admin permissions. And once these are linked, users with the right permission can assign values via set properties mutation, which is a GraphQL API query.
We wrote sustainability analysis results, such as greenhouse emissions, water usage, land usage, back to the design. This helped with collaboration, since once written, the information becomes available to all the designers with whom the design is shared and also makes the design the single source of truth. It also allows the user to view component by component results without constantly switching between applications or between Excel sheet and tabs and so on.
The next is the Design Automation API. Design Automation API provides the ability to use core APIs of your favorite CAD engine, like AutoCAD or 3D Studio Max, Inventor, Revit. It's almost like running a plugin but on call. You can automate your jobs. Say, for example, you want to convert thousands of drawing files to PDF files.
Now normally, you would have to download all the AutoCAD files, run a script on them in the Autodesk desktop environment, and then generate the PDFs and then upload them all back to the cloud. Obviously, there would be a lot of issues. You would be limited by the processing power of your desktop. When you are uploading the output files, you may have network breakage or so on.
Design Automation API for AutoCAD will eliminate this by allowing you to leverage the power and scale of cloud computing. You can also use the Inventor Design Automation APIs for configuring assemblies. You can use iLogic. For Revit, you can use APIs to automate model creation. Documentation. Remember, though, these are not the commands or customization APIs of the host. They are some select operations, not full capability of the plugins.
Now, EcoDesigner supports analysis for both Fusion 360 and Inventor designs. For Inventor designs, direct retrieval of physical properties isn't possible using standard APS services. To overcome this, we utilized Autodesk Design Automation and developed an app bundle which acts as a plugin to fetch these properties.
The app bundle was uploaded to Design Automation, and an activity was created to process the input inventory file stored in the cloud. And by submitting a work item, the DA service generates a JSON file containing mass and material data, which is then again fetched by EcoDesigner for accurate analysis.
Final module, the Autodesk Platform Services Viewer SDK. It lets you create applications to view, share, and interact with design models on your own website from anywhere. Now viewer can display files from AutoCAD, Fusion, Revit and many other formats.
This JavaScript library enables developers to create custom applications that can combine 2D and 3D visualization with business-oriented data. It is an extensible framework that allows you to change the appearance. Controls. Behavior. You can also add your own user experience. We will see that. And it can be leveraged to make project dashboards, digital twins, display planning, and time lining, et cetera.
And we all know the picture conveys a lot more than words or numbers or tables, right? So 3D visualization is a key feature in EcoDesigner, enabling the users to not only visualize the LCA analysis results, but also interact and dissect for better understanding.
We customized the APS viewer to display CO2 emission values from the LCA in a kind of heatmap. The components of your design, which will be responsible for most of the greenhouse emissions, are on the hotter side of the color palette. The ones which are rendered in green are indeed green, eco-friendly components. And the gradient bar on the top provides handy reference to the range of CO2 values.
So our journey has had its own share of ups and downs, and here are the challenges we faced and the lessons we learned. The first obstacle we hit was the material names. Fusion uses its own terminology, and SimaPro has different material nomenclature. There are about 350 standard materials in Fusion that we had to then map to SimaPro's naming conventions in SimaPro's material library.
This exercise was done manually as it was a one time activity. We had to do the same exercise for Inventor, also. Now, I think that this could be a good place where we can use LLMs in future. For now, we just did one time manually.
Second challenge. We are at ease with REST APIs. It's been there for long. So GraphQL was our first. And though it isn't very difficult to understand, but we had to get over our initial blocks that we had, and it was like a learning curve. Also, we were relying on the APIs, especially the manufacturing data model APIs using GraphQL. They themselves were in alpha beta evolving stage, and because of that, sometimes you had to wait for certain APIs to be available. Or at times we had to rework because we sometimes implemented by REST and then replaced by GraphQL and so on.
And the lessons we learned. We learned some good lessons during our journey of making EcoDesigner. The first one, obviously the Manufacturing Data Model APIs offer a simplified gateway to basically read from the design right back to the design, thereby making the design a single source of truth, which is what we ideally want.
We figured that GraphQL APIs are much simpler. They eliminate the need for multiple complicated nested calls required by REST interface. Also, GraphQL provides a very tailored output, which also reduces the need for additional processing of the response.
Before GraphQL, we were using traditional REST APIs, which took two to three minutes for the response. With GraphQL, all the processing is consolidated into one single API call, resulting in a turnaround time, which is five times as fast as the traditional REST API interface. That was huge for us.
So what are our key takeaways? We learned how to use APS APIs to build a web application. How to access the Fusion Data. Process it through another cloud service-- in this particular case, the sustainability analysis. Clearly, it's pretty straightforward, as APS APIs are quite modular with lots and lots of foundational features. Can be done in few days, if not hours.
We also understood the manufacturing Data Model APIs and what you do with it. With its GraphQL interface, makes it very easy to read properties and write properties. Thus, we can make our data model or our design as a single source of truth, which makes the sharing, collaboration, downstream processing way easier.
Many hands have contributed to the making of EcoDesigner. I'm acknowledging the leaders of the respective teams here. Zoe from Autodesk, who is a Senior Strategy Manager and an expert in sustainability solutions. She is a very passionate advocate for sustainability, and she has been a great guide for us.
Caspar is a solutions architect at Pré, And he marshaled the resources there to help us with the technical aspect of accessing SimaPro's LCA services. And finally, Rahul Khande, Project Manager at ProtoTech. He has led the development team and managed the project. And you can see the fantastic outcome of all that.
What's next? It's my call to action. The EcoDesigner is being launched here right at this AU. Now unfortunately, it's being launched at this very moment in another room. You will have to watch its recording on the AU website. Or you can meet Eric Mieras of Pré. He is here with his team, and they are the best experts to guide you on how you can bring sustainability to your organization.
There are lots of initiatives that Autodesk is investing in around sustainability, and they have some amazing solutions as well. If you are interested in learning more about this, please meet Zoe Bezpalko. If you want to know more about the case study that I discussed here, or if you have an idea or an app that you would like to get built, please connect with me or my team. We have a station at the APS zone in the Expo Hall. Now you can shoot your questions. Thank you very much. I hope you enjoyed this session.