Read the full story from Northern Michigan University.
Northern Michigan University associate professor Jeffrey Horn has been awarded $92,500 from the Michigan Translational Research and Commercialization (MTRAC) Innovation Hub for Advanced Computing for a next-generation computing project. Horn received the funding for a Deep-Scale Evolution (DSE) algorithm that he invented that has practical implications for minimizing material waste in industrial production.
“We plan to bring DSE to bear on the real-world shape nesting problems in industry,” said Horn. “Businesses such as automotive manufacturing, textiles and sign making all involve cutting out shaped pieces from a sheet of expensive material, such as steel, glass, leather or titanium. The act of cutting these materials is done by nesting the cutouts close together in an effort to reduce the wasted trim. So DSE maximizes the material usage.
“Our specific goal is the development and deployment of DSE on a powerful server specially designed to run DSE with massive population sizes yet short computation times. To do this, we’ll take advantage of the latest revolution in processing power—the graphics processing unit or GPU—originally developed to render 3D graphics for games and animations but now used for massively parallel algorithms, like Deep Learning, Bitcoin mining, and simulated evolution like DSE.”