Freudenberg Sealing advances AI use in manufacturing, testing

WEINHEIM, Germany —While reliable quality control is essential to the smooth-running of manufacturing operations, products are sometimes rejected as being defective even though they are actually problem-free.

To reduce this often costly source of scrap and inefficiency, Freudenberg Sealing Technologies (FST) has turned to the use of artificial intelligence (AI) for automatic visual inspection processes at some of its plants.

The technology was initially introduced through a pilot project at the company's Oberwihl plant in southern Germany, where it reduced scrap during final inspections by 50 percent, the company reported.

AI has allowed FST "to better differentiate between subtle differences, said senior engineering specialist Helmut Hamfeld. "We can more easily discover whether the rejected product actually deviates from the norm, or whether a shadow has produced a false result."

FST has since introduced AI-optimized, automatic visual inspection systems at other facilities, including for an application during production on a molding machine its North Shields facility in the United Kingdom.

At the north of England plant, the system recognizes whether the mold, or the cavity into which the sealing material is pressed, is free or still occupied. When an alarm sounds, the part still in the mold can be removed quickly.

Machine downtime has reduced "significantly," Hamfeld reported. "We are identifying errors as they come up and not at the very end of the production chain, (leading) to less mold damage and reduced the need for replacement parts."

Another positive side effect is that production processes can be accelerated, the engineering specialist continued, noting that due to this benefit the system is set to be introduced at FST plants in Spain, Mexico and Turkey.

Meanwhile, having developed extensive experience in the use of AI in automatic visual inspection, FST now is turning to machine controls to test an AI application that involves the use of up to 3,500 measuring points and sensors to capture the heating time or pressure conditions in seal-production processes.

Still at the pre-implementation stages, the system is designed to provide immediate prompts to adjust the parameters where the data enters a range that does not guarantee a good result.

According to FST, this capability would be a step toward achieving "a zero-error, automated process chain that would make final quality inspections unnecessary."

"We are aware that we now can solve a great many problems with AI that we were unable to solve before," said Hamfeld. "But AI is not a silver bullet; we are still seeing that our engineers have such specialized knowledge that we are unable to configure it into an AI application."

Gaining an understanding of how AI makes its decisions continues to be a challenge as well, according to the engineering expert, while noting that FST's advanced analytics team is making progress in this area.

"We are now using algorithms that allow us to check the correlation underlying AI with regard to causality," Hamfeld said. "In any case, the advantages of AI use at Freudenberg Sealing Technologies are uncontested."