Machine learning and software cut waste from 3D printing

The process, which predicts shape deviations for all types of 3D printing, has been dubbed ‘convolution modelling of 3D printing’ and is described in IEEE Transactions on Automation Science and Engineering.

How collaboration is driving advances in additive manufacturing

Additive manufacturing can reduce industry’s environmental impact

"What we have demonstrated so far is that in printed examples the accuracy can improve around 50 per cent or more," said team leader Qiang Huang, associate professor of industrial and systems engineering, chemical engineering and materials science. "In cases where we are producing a 3D object similar to the training cases, overall accuracy improvement can be as high as 90 per cent."

"It can actually take industry eight iterative builds to get one part correct, for various reasons, and this is for metal, so it's very expensive," Huang said.