Leading the advance of global manufacturing

Dec. 12, 2018

By Tona Kunz

LEMONT, IL—Argonne and MIT look to expand how high-performance computing can advance the manufacturing ecosystem.

Manufacturing is crossing borders and traditional concepts of supply chains by using virtual methods and custom fabrication capabilities.

Whoever owns this new process for making products will lead in global manufacturing. The United States is poised to lead this next wave of manufacturing innovation by leveraging the national laboratories and their unique tools and expertise.

The Manufacturing Science and Engineering initiative at the U.S. Department of Energy’s (DOE) Argonne National Laboratory is beginning this process by connecting with innovators in this domain, working toward providing access to predictive models and large materials data sets connected to supercomputing facilities. These resources can save energy and prevent materials waste as many users start to build their own products and participate in the innovation marketplace through the democratization of manufacturing processes down to local communities.

That’s the vision of Santanu Chaudhuri, director of Argonne’s Manufacturing Science and Engineering initiative and a professor at the University of Illinois at Chicago. He outlined a possible path to new uses for supercomputers and multiphysics models and examples of current work in that direction during a keynote presentation at the ACM Symposium on Computational Fabrication at MIT.

MIT’s Center for Bits and Atoms provides education support and open-source software through the Fab Foundation, a global network of entrepreneurs and innovators that use fabrication laboratories – also called ​Fab Labs.” The Fab Labs provide kits that allow users to achieve industrial-grade fabrication and use tools such as laser cutters and 3-D printers, and open up the ability to make new products and prototypes using cost-effective methods. There are around 1,200 Fab Labs in the world, and the number has been doubling about every 1.5 years.

In Chaudhuri’s presentation, ​Rethinking Fabrication in a Continuum between Edge and Petascale Supercomputing,” he provided examples of how an adaption of the current high-performance computing analysis and data delivery model built by researchers at Argonne and universities can impact local entrepreneurs and innovators. Traditionally, access to supercomputers at national laboratories has been limited to professional researchers. But barriers to access are being removed by developing collaborations between national laboratories and leaders in Fab Lab design, such as MIT. Design of new Fab Lab technologies and access to the national laboratory systems’ supercomputers and materials characterization and testing can help accelerate innovation.

These types of collaborations will give researchers with little knowledge of supercomputing methods the ability to iterate and predict new designs without having to spend years in trial-and-error prototyping. In slightly more complex cases, advanced computing and a materials genome approach can help researchers design new materials that can support a more complex product design with new attributes, such as increased strength or reduced emissions.

New advanced manufacturing techniques, such as 3-D printing, open the door to the development and design of new materials and products. But to take full advantage of these techniques, innovators need to better understand the unique, complex micro and macro structures these techniques create in materials.

That is where supercomputers come in. Because they can ​crunch” immense amounts of data about variables in materials and manufacturing processes, supercomputers allow users to more accurately predict product lifetime and performance. Petascale supercomputers can analyze more than one quadrillion floating-point operations, or mathematical equations, per second. To make these data useful for real-time manufacturing, though, they need to be paired with edge computing, which lets innovators use cloud computing and sensors on manufacturing equipment to collect and analyze data in real time closer to the product development site. Argonne is building such prototypes to reach a much broader audience through Fab Labs.

Combining Argonne’s analysis power in petascale and exascale computing — a quintillion calculations per second—with edge computing on the factory floor and in Fab Labs could revolutionize the way communities train the next would-be entrepreneurs in manufacturing science, advanced materials and product design. Argonne has made it a key lab mission to partner with universities such as MIT to focus on this new challenge and impact this rapidly growing field.

Argonne National Laboratory has the complete article

Sponsored Recommendations


To join the conversation, and become an exclusive member of Electronic Design, create an account today!