GPU-powered manufacturing at Foxconn Interconnect Technology

Foxconn Interconnect Technology (FIT) is rolling out artificial intelligence (AI)-powered inspection systems in its injection molding manufacturing lines. The move is likely to bolster production throughput and improve the quality of devices.

Source: NVIDIA blog post.

AI is used for automated, high-precision inspection of components and tools. The manufacturing giant aims to address a shortage of trained inspectors with this technology, aiming to eventually free up these expert technicians for more valuable quality assurance positions. FIT plans to expand the initiative to 30% of its production lines by 2019.

FIT has trained convolutional neural networks to identify defects in 10 different categories of components. The company’s data scientists initially fed its neural nets 600 images for each defect. The networks continue to improve the system from new images captured by industrial cameras.

NVIDIA HGX and DGX systems helped handle FIT’s requirements — hundreds of terabytes — for training its AI inspection models in the company’s private cloud. FIT also installed a DGX Station onsite for ongoing training and inference as well as the Jetson TX2 module to widen its inspection coverage reach at lower price points.

“Through NVIDIA’s outstanding computational capabilities, FIT’s AI inspection capability was able to reach more than 95% accuracy, approaching the level of quality and efficiency of professional human inspection,” said Joseph Wang, CTO, FIT.

FIT’s CPU socket injection molding machines are fitted with five cameras to spot 1 mm defects among the millions of flawless components per day produced by its precision manufacturing processes. This application of deep learning was not possible with CPUs, and instead required GPU acceleration.

FIT has ambitious goals for its AI. It eventually plans to use the technology to maximise yield quality, boosting the finished production output.

“We are aiming for fewer returns. Quality is the main goal — this is the first step of a much wider plan,” said Wang.

Smart factories can nearly double the profit margin of the average automotive manufacturer, estimates Capgemini. More than half of executives at large manufacturers surveyed by the consultancy said they have invested US$100 million or more on smart factories over the past five years. 

FIT's parent Foxconn manufactures 40% of consumer electronics worldwide. Foxconn Chairman Terry Guo has said publicly that the AI programme represents the company’s opportunity to “transform from traditional manufacturing.” The efforts are positioned to reduce costs, create efficiency and deliver improved results for its customers.

Explore:

NVIDIA founder and CEO Jensen Huang will kick off CES with a keynote on January 6, 2019 in Las Vegas, the US, at 8 pm (January 7 12 pm).