Skip to content

AM, Big Data, VR, AR all crashing DFM’s party—in a good way

By Sean Lyngaas Contributing Editor, SME Media

Change coming quickly to ‘design for manufacturing’

Design for manufacturing has been around for decades, but industry insiders say the next few years will be critical as technologies like additive manufacturing (AM) and virtual reality (VR) shape the future of the industry.

FabriceAgnesDassaultSystemes-300x169.jpg
“It is a somewhat untapped field, but using big data for DFM offers “a huge return on investment,” says Fabrice Agnes, senior director, CATIA portfolio, at Dassault Systèmes.

Design for manufacturability (DFM) software allows designers to check their part models up front for inaccuracies and inefficiencies, thus saving them money. The goal is to get closer to the designer’s original intent up front without having to make costly revisions. Over the years, DFM has gotten more accurate and responsive to a designer’s needs.

“At the design stage now, with very little information, we can come up with the cost estimate for an injection molded part, let’s say, that replaces the function of three sheetmetal parts with the various fastening operations,” said Nick Dewhurst, executive VP of Boothroyd Dewhurst, a Rhode Island-based design for manufacture and assembly (DFMA) specialist. “And we can decide whether the tooling investments justify the move in that direction.”

“People have latched on to DFM as a really good way to negotiate with suppliers around cost of individual parts,” he added.

Progress in DFM notwithstanding, parallel advances in AM (aka 3D printing) and VR are driving home the point to designers that so much more can be done with DFM. And while it may be hard to find a manufacturer unfamiliar with DFM, there are still plenty that have been slow to adopt it, said Fabrice Agnes, senior director, CATIA portfolio, at Dassault Systèmes.

“It takes decades, depending on the industry, to move forward from a traditional way of thinking to a new approach,” he said. “The early adopters are definitively, at the moment, aerospace and automotive.”

One of the building blocks of recent advances in DFM is known as generative design.

The concept is simple, but powerful: a designer enters design goals for a part, along with variables, such as cost constraints, into software that generates an array of possible designs. The software can learn from each query so that it generates more useful design options with more queries.

“It’s a complete revolution of the way we are designing things,” Agnes said of generative design, referring to an ability to generate new shapes based on detailed study of design constraints.

The goal, said Diego Tamburini of Autodesk, is to use generative design to find a model that can be manufactured right off the bat, rather than hashing through countless choices.

Autodesk, a 3D software firm, is very bullish on generative design and would like to expand its use as widely as possible, from the factory to down the supply chain, he said, noting that Stanley Black & Decker uses Autodesk software to halve the weight of crimping parts.

3D printing fuels the imagination

While generative design’s intrinsic value to DFM is assured, AM is still up in the air. 3D printers have become cheaper in recent years, making the technology more appealing to manufacturers of all stripes.

“I think that we are reaching right now the inflection point where 3D printing is kind of migrating from those use cases that are used only for prototyping into production,” Tamburini said.

However, while no one doubts the promise of additive manufacturing, not everyone is convinced it is ready for primetime.

Dewhurst said his firm hasn’t invested big resources into AM because a standard production process for the technology has yet to emerge.

“We wouldn’t want to put a lot of time and effort into developing cost models for something that ends up being not the process people use,” he said.

Practitioners are trying to change that, in part through a consortium to advance AM known as 3MF. Microsoft, Autodesk, and Dassault Systèmes are all on the consortium’s board.

“3D printing today is very fragmented,” said Kishore Boyalakuntla, senior director of product portfolio management at Solidworks Corp., which Dassault acquired in 1997. While many big companies are already excelling at AM, others are just entering the field, so harmonizing standards will take time, he added.

New skill sets needed

The advent of AM is forcing engineers and manufacturers to develop different skillsets, practitioners said. Cutting-edge DFM is a completely new way of designing things, and many companies lack the internal skills to exploit these new technologies, Tamburini said.

Some of the wild shapes generated by 3D printing are simply foreign to designers, he added.

“A human brain probably cannot even come up with those designs. That’s why we’re looking for software algorithms that mimic the way that nature designs things,” Tamburini said. “So it is a completely new set of skills, new possibilities that are not automatically picked up. There is a learning curve that is still steep for some.”

At the same time, however, manufacturing has become more accessible to countless companies as they buy 3D printers.

One Solidworks customer employed a mechanical engineer just two years out of college who was designing and cutting 1000 different parts for the company, Boyalakuntla said.

Part of the learning curve is getting manufacturers to embrace the notion that designs should be checked for manufacturability up front, said Paul Brown, senior marketing director at Siemens PLM Software.

“There is a cultural barrier. Lots of times, people think that the checking should just come at the end,” he said.

Dewhurst reckons that 80% of his company’s customers do not use DFMA in the early stages of the development of new products. An entrenched status quo in which designers cling to their ways of doing things explains this predilection for not using DFMA up front, he added.

“People have to kind of take a leap of faith and know that the investment in time during the design cycle will pay off for them later.”

Big Data untapped

As in other industries, the data emanating from a surge in connected devices—the phenomenon commonly known as Big Data—presents an invaluable opportunity for designers to deliver a more precise and customized device.

And industry insiders say designers have only skimmed the surface of Big Data’s potential.

The next phase for Brown’s team is figuring out how its members can “tie the kind of information that’s coming back from my machine… on the factory floor back into giving me things that help advise the designer, help advise the manufacturing engineer as they’re planning how to manufacture a part,” he said.

It is a somewhat untapped field, but using Big Data for DFM offers “a huge return on investment,” Agnes said.

He pointed to Dassault Systèmes’s acquisition of Exalead, a data platform company, in 2010. Exalead has an application, One Part, that can calculate the cost of a single part in a long list of them in less than a second, he said. This level of precision is increasingly needed in an industry in which a manufacturer’s supply chain can involve dozens of different suppliers.

VR, AR seen adding promise

Looking further into the future, VR and augmented reality (AR) can propel DFM to new heights in a matter of a few years, practitioners said.

To that end, 3D glasses could be popping up more and more on the factory floor.

VR brings much more accuracy in 3D perception compared with a simple flat-screen display, Agnes said. 3D glasses give you depth perception that can, for example, help a designer see exactly how a wire needs to be placed to connect different components, he added.

Glasses give you depth perception, which helps with product definition validation so an electrical designer can see exactly how the wire needs to be placed to connect the different components.

“Virtual reality is more than only a gadget for gamers,” Agnes said. “It’s really a change in the way people perceive and make the right decision about the way to design things.”

Siemens’s Brown echoed that point and emphasized that VR can help with the assembly process by zooming in on a specific part and gleaning more information from it. Practitioners can quickly learn more about the material and machine origin of the part with VR, he added.

The next few years will be critical for a DFM field that has steadily matured over the years but that some manufacturers are only beginning to embrace.

How the uncertain-but-promising use of AM pans out could open new doors in imaginative design. And the greater use of VR, which can deliver the accuracy and visualization of a part that designers have long craved, could transform the field.

  • View All Articles
  • Connect With Us
    TwitterFacebookLinkedInYouTube

Related Articles

Always Stay Informed

Receive the latest manufacturing news and technical information by subscribing to our monthly and quarterly magazines, weekly and monthly eNewsletters, and podcast channel.