Do you remember when the first simulation based optimization tools hit the industry? From what you might have read, it might have seemed as if nirvana had been found. All you have to do is define a simulation, define the design variables that should change and identify your objective and poof, the perfect design would emerge. But reality more than tempered that vision. As some organizations started to use the technology, they found it required a good bit more hand holding than they first imagined.
Fast forward a good number of years and I sat down to get an update from the folks at Altair Engineering on their products. Going in, I certainly didn’t think the one thing that struck me as most compelling would be optimization. But it was. What follows is a little background on their optimization technology, what capability it provides as well as my own commentary and analysis.
In the early 1990’s, Altair Engineering exploring university research on topology optimization. After several years of writing commercial code, this effort culminated in the release of their OptiStruct software in 1994. Initially, OptiStruct allowed for stiffness maximization with volume and model constraints. Over time, a variety of types of analyses as well as analysis responses were were added as part of the optimization process as objectives or constraints. Today, OptiStruct is part of Altair Engineering’s Hyperworks suite as well as solidThinking Inspire, a conceptual design tool.
Optimization is all about varying parametric dimensions of a feature-based CAD model to get different iterations on geometry, right? Well, when it comes to Altair’s OptiStruct, there’s a little more to it than that.
Optimization for Concept Design
So what exact does Altair’s OptiStruct do differently? Well, it actually suggests potential shapes for your design. Let me explain.
Let’s say you are designing a part to carry a structural load. With OptiStruct, you essentially model the outer boundary that this part could occupy. From there, you add in your normal constraints and loads. After that, you define your objective and design constraints. From there, OptiStruct essentially iterates on removing material in various locations to meet the objective and stay within your design constraints. Essentially, by performing a structural simulation on the block, OptiStruct understands which material within the block is and is not carrying the load. From there, it’s not to hard to see how OptiStruct could remove the material that isn’t carrying the load and remove it.
At first glance, it may sound fairly unimpressive, but what is key is in defining the outer boundaries that the part could occupy. That starting shape could be as simple as a block or cylinder. But realistically, it could be far more complex. Perhaps you need a part to fit within a number of other assemblies. You could potentially use a CAD model to extract the unfilled volume that the part should occupy along with its attachment points. That becomes the starting point for which OptiStruct could start removing material and tell you what the conceptual shape for the part should be.
Optimization for Detailed Design
Now the example above is interesting but it is also very simplistic. Now, expand it to take into account natural frequencies and temperature variations and it gets far more impressive. The good news is that OptiStruct can go beyond structural analysis with its optimization capabilities. It covers a wide range of physics including:
- Modal Dynamics
- Kinematics and Dynamics
All of these types of characteristics can be taken into account when executing this kind of optimization.
Commentary and Analysis
OK. So it’s interesting technology. What’s it mean?
Functionally Driven Concept Design
Have you been tracking the new buzz about concept design in the industry? It’s actually quite compelling. Some say, me included, that CAD tools are very frequently used to document the product after most design decisions have been made. Why? Because CAD is often very difficult for engineers to use, especially when they are running from their desk, to the shop floor, to the prototype shop and the test lab. There’s little time for them to specialize in complex software like CAD. However, the emergence and acceptance of direct modeling has changed that story. Direct modeling lets engineers just grab and drag geometry without having to fiddle with parametrics or features, which can often times need to be handled with care. In addition to that a number of CAD software providers have started offering new tools purpose built to address the design of concept design. With a combination of 2D sketching and direct modeling, it is far easier to capture ideas and concepts in digital models than ever before. But in the midst of all that, there’s one caveat to the whole thing.
A good number of parts are almost purely functionally driven.
Now, don’t get me wrong. Some parts need to be aesthetically pleasing. Some parts are simple and straightforward. However, have you ever looked at the assembly of a car door? What about the inside of an airplane’s wing? Every nook, cranny, divot, rib and whatnot serves an engineering purpose. Practically no one cares what it looks like. It just needs to serve its function. And ultimately, that’s why I find a tool like OptiStruct terribly compelling. You can set it up and let it go. Ultimately, you can get to a better design.
But there’s another really interesting reason to consider this kind of technology. What value do engineers add? If a technology like this can let an engineering flip it on, walk away and work on another design, it is making them more productive. Obviously, they need to validate the results and the design. But I think there is a hidden productivity advantage hidden here as well.
So, it’s all roses? Not quite. While considering this technology, I had to ask myself: how frequently do you use this technology? Would it be every part?
I think it is fairly obvious the frank answer is no. Although there would be some advantage, I think you start approaching the point of diminishing return. You might have an optimized product but you may not stay on schedule. This stuff can take some time. But can you strategically use it for the parts that have a bigger impact on the cost, manufacturability and performance of the product? Absolutely. And I think that is where it would shine. Use it to do the grunt work on big complex parts that impact the design while engineers use some of these newer purpose built CAD tools to buzz through the concept design of parts that simply need to get done. That feels like the right combination of technologies to me.
One other point that needs to be taken into account is the importance of getting to an optimized design as opposed to a feasible design. So many engineers are rushed with their workload that, even if they wanted to, they may not be able to spare the time to take this approach with the impactful parts. I think if this technology is adopted, and the advantages and benefits warrant it, that engineering organizations should triage parts early on into different design processes. The parts that heavily impact product traits should use something like OptiStruct. All others take a abbreviated route.
Conclusions and Questions
Well, this is a new type of technology we haven’t looked at before. Let’s recap.
- historical summary of the product
- OptiStruct lets users model the space that a part could potentially fill, the starting volume, and then figures out which materials affects design objectives and constraints. It removes material that doesn’t make a different, leaving the shape representing the concept design.
- The shape of the starting volume can be very complex, potentially being created from an empty air extracted volume within a CAD assembly model.
- This technology represents a significant value add for parts that are functionally driven and are impactful on the product’s key characteristics.
- Adoption of this technology should come with some procedure to triage parts into a process where concept optimization is used or not.
Alright. Those are my thoughts. Here’s my questions for you. Do you think optimization has gotten a bad rap in the past? Have you had to overcome objections to its use in your organizations? If you use optimization, how do you determine which designs get optimizations and which do not? I’m curious to get your perspective.
Take care. Talk soon. And thanks for reading.