The term generative design has been popping up in the manufacturing world of late. Its promise is to create many design permutations to let engineers choose an optimum one that meets sometimes conflicting requirements.
The problem is that many of the designs that result from using generative algorithms are complex, even organic looking, with complex curves and webs. This has led to a strong association of it with additive manufacturing (AM), which can be expensive to obtain and operate. Materials are more limited compared to traditional machines. This link has discouraged the idea of generative design for many machine shops.
However, traditional shops do not need to invest in new AM equipment to build parts inspired by generative techniques.
Generative designs are created by software that uses real-world manufacturing constraints and performance requirements to design a number of permutations. The final results are usually in the form of a CAD-ready model. They balance conflicting requirements to varying degrees, which might include, say, minimum weight and maximum safety, or minimum cost and maximum functionality.
Generatively inspired designs are created through a generative process but are further augmented to fit into the manufacturing capabilities of a specific shop. With the advancement of generative design software, a manufacturer open to new trends can more easily bring a generatively inspired product to market.
At the core of generative design is the development of software tools capable of taking on an active role as a co-creator. They automate exploring solution variations based on supplied parameters.
“You might have flexibility to look at your design as aluminum or steel or multiple grades of steel based on its performance requirements and the ability to get those sorts of materials,” said Mike Smell, product manager of generative design in Fusion 360 at software developer Autodesk Inc., San Rafael, Calif. “And then you can say, if we are going to make a net new product maybe we have flexibility in the way we do this, from a material or manufacturing process point of view.”
From there, the program provides a number of solutions that give the person using Autodesk’s system the ability to conduct a trade-off analysis. “This shape might be the lightest, this shape may be the cheapest, and this shape may be between all of those,” he said, referring to three possible designs the software would recommend. It could be many more. “The designer or engineer then looks at all of those and determines which one is most appropriate.”
Autodesk is in the process of making generative design more accessible to mainstream manufacturing. The company has started looking at traditional mills for fabricating 3D parts from generative designs. “In Fusion 360 today, we [Autodesk] also have support for three- and five-axis milling methods for generative design,” Smell said.
This option of manufacturing may be slower than 3D printing given the fixturing time and tool changes associated with traditional milling. On the other hand, milling may be more cost effective for some shops dabbling into the prototyping of generatively designed parts since the shop doesn’t incur the cost associated with bringing an industrial-sized 3D printer onto the shop floor. That is not to mention the cost-per-part that occurs when using spools of printer filament.
Additionally, the option for 2D cutting with technologies like waterjet and plasma are being examined for their cost-effectiveness. “We are actively working on a number of solutions on the path toward enabling mainstream manufacturing methods: two-axis or 2D cutting, two-and-a-half axis milling machines, and a number of other manufacturing methods,” Smell said.
Though these 2D pieces would be more traditional looking, the outcome could still result in parts that are lighter and stronger than the originals. Given the dramatically lower cost of cutting with traditional 2D methods, it is feasible to see waterjet, plasma, EDM, and laser used in shops experimenting within their own boundaries with generative design.
There is common agreement that if generative design software development is to progress and be accessible and viable to mass market manufacturing, additive manufacturing needs to cost less—or users need to adapt it to traditional manufacturing methods.
In 2018, Phillip Keane wrote an article titled “Generative Design: The Road to Production” for Engineering.com. “…while generative design is driven by additive manufacturing, it is not exclusive to 3D printing, and companies involved in casting could stand to benefit from generative design (as long as they modify their mind-set and workflow accordingly).” As Keane pointed out, for generative design to be accepted by a manufacturing base, there will need to be a shift in conventional mindsets as well as workflow considerations.
We have a tendency to believe that solid pieces of material impart an aesthetic of strength and reliability. Furthermore, when a customer orders a part, they have a belief in what the end part will look like. It is true that generatively designed pieces have an aesthetic that isn’t always the most customer pleasing. Often the outcome of generative design is described as being organic or even alien.
There is also a mindset around generative that believes the cost of 3D printing generatively designed parts is an expense not worth taking on. The shift in thinking comes in realizing that a machine shop doesn’t need to invest in large-scale 3D printers to benefit from the generative design part creation. As discussed above, there are software developments enabling the conventional milling of generatively inspired parts.
The workflow shift will affect every shop’s machining processes. As Smell discussed earlier in the article, steps are being made to allow for traditional machining tools to create generatively designed parts. But to get the result you may be looking for, a shop may have to rearrange its order of operation to account for the possibility of added machining time. Near-net cutting may become much more common, or even necessary, when the facets of the new generatively designed piece takes hours to mill.
Keane argues that the end effect of generative designed or generatively inspired parts will eventually be a net gain once the concept has been fully adopted, rather than experimented with, by a shop. “It seems to be a case of how you adapt the generative design tools to be used with traditional manufacturing techniques,” writes Keane. “After all, if you use less material and still manage to reduce the number of tooling operations, the unit cost will still be reduced. Generative design is providing designers with new ways of looking at old problems.”
There are machine tools other than 3D printers that might be perfect for generatively designed or generatively inspired parts. Since generative design often looks at the possibility of stronger, cheaper, lighter, and faster ways to produce parts, there will be a consideration of materials. Near netting and 2D cutting are easily achieved with more traditional machine tools like laser, plasma, EDM, and waterjet. Of those three, waterjet is the only tool that allows for a wide variety of materials to be cut.
Carl Olsen, director of software development for Kent, Wash.-based OMAX Corp., said, “The tools are there right now, and they’re good [when used] with the software and the traditional manufacturing machines that can cut arbitrary shapes: waterjet, plasma, EDM, and traditional milling.”
With the exploration of newer designs, the possibility of traditionally manufactured parts moving away from aluminum and steel in exchange for composites is looming. In that climate, a versatile tool is needed. Whereas plasma, EDM, or laser have their restrictions as to what they can cut, abrasive waterjet is a near-universally applicable cutting tool.
“If it’s flat, using a waterjet would be far faster than using additive machining and there will be a much better selection of materials to work with,” Olsen continued. “Your 3D printer may be good at printing steel but not so good with titanium or something else.” With waterjet, material considerations widen to include nearly everything under the sun.”
However, as the onset of generatively designed and generatively inspired parts affects machine shops around the world, what is considered the “right tool” for the job will change. 3D printers are still a ways off from being financially viable in major manufacturing. With the tools you already have, generative design and generatively inspired parts can be an encouraging direction in the evolution of part machining. At the very least, it can be an option when deciding how best to manufacture a new part with the tools on hand.