Written by Rob Cohee, Autodesk
Here comes the part where one software vendor says "no, we have the best new features!" and the other one goes "no way we do!", then the other one says "haha we're so confident our users love us we can do whatever we want to our subscribers, we don't need any stinking new features!" This is not going to be another one of those posts and if you expected that from me, you haven't read enough of my musings.
Why do I say this is huge? Think about the way we model our designs for a minute. No, not the picks and clicks but fundamentally what are we actually doing when we are modeling? We are documenting either what already exists or giving a design our best guess. Best guess? Yes, early in the design process the form of design is quite frankly a guess. An educated guess mind you, mostly based on experience that the design is going to hold up to the stresses that are going to be applied to it. Some of you will then simulate those forces using FEA, most designers skip that part and go right into production. 'Cause ain't nobody got time for that!
The way we design today is an iterative process. Let’s Google that for a moment shall we?
- iteration -
Repetition of a mathematical or computational procedure applied to the result of a previous application, typically as a means of obtaining successively closer approximations to the solution of a problem.
We take a design, we manufacture it, we try it and if it works great, if it doesn’t we try and try again. One of the most common methods to avoid this try and try again iterative design process is to over–engineer our designs.
A little background on me – the bulk of my design experience was while I worked for Union Pacific Railroad where my job was to investigate safety-related issues and come up with new processes or equipment that would make repair or maintenance processes safer. The weight of my designs were never a consideration, and from an aesthetic point of view it had to look as tough and as durable as the guys using it. When quarter inch thick material would do, the rule of thumb was to bump it up to three eights of an inch just to be on the safe side. Does this sound familiar?
Back to the idea of turning things upside down. What if we could better predict the form our components need to be based on the functions they perform? What if you could get closer to the very best solution by simply telling your design tool what the component will do in the design? What if we could take some of the time consumed by this iterative process, and spend the right amount of time on delivering an optimised solution?
- predictive -
Relating to or having the effect of predicting an event or result.
Predictive design takes the try, and try again or more commonly the “just to be on the safe side” aspect of the design process and significantly reduces the amount of time it will take to achieve an optimised solution.
So what’s the big news?
The Autodesk Inventor team will soon release new functionality that we are calling Generative Design and it’s the most exciting thing I think we have ever put into any of our products. It takes this idea of predictive design and makes it a reality. Its functionality that is going to take so many cycles out of the current iterative design process that it will simply become the way we model going forward. It puts the “A” in CAD. If that sounds corny, blame my colleague Sam Antos, if it sounds cool… well it was still his idea.
Lets take a look into Generative Design. In this example we have an arm of a clap that is going to have a heavy force applied to it. Start with the material of course, which in this case is steel.
The setup for this is very simple. Set the constraints, pin above and frictionless below.
Kick in Generative Design, and Autodesk Inventor gives you an optimum shape based on the inputs provided.
If you're thinking "That's awesome!" Yes, yes it is. And I had the same reaction along with high fives, a few Woot! Woots! Bubb Rubb style, and a happy dance all in one. Yes I knocked over a table, might have broken a conference phone but it was worth it.
Now what can I do with this shape? Do I have to now 3D print this thing? No, that's the misconception here with this organic shape. You turn this shape into something that you have the capacity to manufacture. Now of course you could 3D print it if you wanted to because Inventor has some impressive 3D printing capability, but it's your choice - is it a weldment? Are you going to nest that thing and cut out dozens at a time on a table, I don't know... That's the thing you spend less time fussing with the shape and more time optimizing for your purposes.
Now we're going to be talking about this a lot over the next couple of months. Because a) It's awesome, and b) I get asked all the time - "Are you guys putting everything into Fusion 360?" No, were putting everything into our manufacturing portfolio. We're building our portfolio on a vision we are calling The Future of Making Things. Right now you still know us as a product based company which is why many of you ask me where we are going with Inventor. Over time you'll hear us shift how we talk about our technology. We're going to be more focused on solving design challenges and quite honestly the various products as you know them by name will be secondary to what they seek to achieve.
With Generative Design we've focused on enabling engineers to get closer to the optimized form more quickly. Generative Design functionality will be included in a mid-year release of Autodesk Inventor 2016 'round the end of October 2015. The advantages here are decreasing your time to market, reducing the amount of material used to produce, and lowering overall shipping costs just to name a few. You're not in the business of being "CAD Experts", you're in the business of manufacturing great products.
Contact Redstack for more information on Autodesk Inventor and generative design.
Originally published by Rob Cohee - See https://www.linkedin.com/pulse/autodesk-inventor-turn-your-world-upside-down-rob-cohee