Raed Al Kontar, University of Michigan Industrial and Operations Engineering (U-M IOE) Assistant Professor and Chinedum Okwudire, U-M Mechanical Engineering (ME) Professor have received a grant from Cisco Systems Inc. to develop novel federated learning algorithms that address descriptive, predictive and prescriptive machine learning opportunities in large-scale distributed 3D printing.
“We are excited about this opportunity to continue to collaborate with CISCO to leverage machine learning and distributed computing to advance and democratize 3D printing,” said Okwudire.
The researchers plan to implement their methods on Cisco’s Flame Platform, which allows users to set up a variety of distributed computing architectures for federated learning. 3D printing companies will then be able to use this technology to streamline their workflows creating a more efficient business model.
“The overarching goal of distributed 3D printing is to produce parts faster, better, and cheaper,” said Kontar. “Through federated learning, we aim to allow dispersed entities to collaborate while preserving their intellectual property.”
The project titled “Federated Learning for Large-Scale Distributed 3D Printing built on Flame” will be completed at the end of 2024.