University of Illinois Urbana-Champaign engineers build system to trace additive manufacturing fingerprints.
Traceability is crucial in manufacturing, supporting quality assurance and quality control, sustainability, regulation, and security, to name just a few essential considerations in virtually any industry. Unfortunately, it’s not always easy to trace a part’s history.
In bygone days, the mark from a blacksmith or a cobbler on a sword or a shoe could be sufficient to identify where that object came from and by whom it was made. But the industrial revolution took us away from those easily traceable fingerprints by replacing handcrafted pieces with machine-made products.
Today, we may be coming back around to this sort of “easy” traceability but via a much more complex route that involves 3D printing and artificial intelligence (AI). Engineers at the University of Illinois Urbana-Champaign have constructed an AI system that can detect the signature of the specific 3D printer used to build a part from photographs alone.
“We are still amazed that this works,” said Bill King, a professor of mechanical science and engineering, in a press release. “We can print the same part design on two identical machines –same model, same process settings, same material – and each machine leaves a unique fingerprint that the AI model can trace back to the machine. It’s possible to determine exactly where and how something was made. You don’t have to take your supplier’s word on anything.”
It’s not difficult to imagine the potential benefits this technology could have for supply management and quality control. Any engineer who’s spent enough time working with suppliers knows that they can make changes to materials and/or machines without notice, occasionally with disastrous results.
“Modern supply chains are based on trust,” King said. “There’s due diligence in the form of audits and site tours at the start of the relationship. But, for most companies, it’s not feasible to continuously monitor their suppliers. Changes to the manufacturing process can go unnoticed for a long time, and you don’t find out until a bad batch of products is made. Everyone who works in manufacturing has a story about a supplier that changed something without permission and caused a serious problem.”
The insight that drove this research came from a study on repeatability, when King and his team noticed a strong correlation between dimensional tolerances and individual machines. Investigating further, they discovered the presence of so-called “fingerprints” in the parts that could be used to determine the particular material and machine used to make a given part.
“These manufacturing fingerprints have been hiding in plain sight,” King said. “There are thousands of 3D printers in the world, and tens of millions of 3D printed parts used in airplanes, automobiles, medical devices, consumer products, and a host of other applications. Each one of these parts has a unique signature that can be detected using AI.”
King and his team developed an AI model to identify production fingerprints from photographs taken with smartphone cameras, using a dataset of photographs of 9,192 parts made on 21 machines from six companies using four different fabrication processes. When calibrating their model, the researchers found that a fingerprint could be obtained with 98% accuracy from as little as one square millimeter of the part’s surface.
“Our results suggest that the AI model can make accurate predictions when trained with as few as 10 parts,” King said. “Using just a few samples from a supplier, it’s possible to verify everything that they deliver after.”
The research is published in the journal Advanced Manufacturing.
