At its in-house INTECH trade fair, TRUMPF is  showcasing its Cutting Assistant, an application which uses artificial intelligence to improve the quality of laser-cut edges. To use it, production employees take a  picture of their components cut edge with a hand scanner. The solution then  assesses edge quality by evaluating for objective criteria such as burr formation.  With this information, the Cutting Assistants optimisation algorithm suggests  improved parameters for the cutting process. The machine then cuts the sheet  metal once more. If the quality still does not meet expectations, the user has the  option to repeat the process. 

“With the Cutting Assistant, TRUMPF is once again  demonstrating that we are a pioneer in the use of AI in industrial manufacturing.  Users aren’t just empowered to combat the skilled worker shortage, they’re also  able to save time and money. This gives them a competitive edge when it comes  to productivity,” says Louisa Peters, Product Manager, TRUMPF. 

The high-tech  company offers this solution for all TruLaser series featuring a power output of 6  kW or more. Most owners who are already in possession of TRUMPF machines  should be able to simply download the Cutting Assistant. 

During laser cutting, users often have difficulty determining the correct  parameters for their material grades. Materials which are not optimised for laser  cutting often produce edges with wide variations in cut quality, forcing production  employees to constantly change the technology parameters. This entails  adjusting each individual parameter one by one,  a process which demands a lot  of time and employee experience. This presents challenges for many companies,  especially those with inexperienced workers in production positions. By  integrating the Cutting Assistant into the machine software, optimised parameters  can be transferred seamlessly into the software without the need for  programming. This saves time and reduces errors.

TRUMPF experts have cut thousands of parts and drawn upon their years of  expertise while developing the Cutting Assistant, using their extensive knowledge  to train the softwares algorithm. For TRUMPF, the work on the Cutting Assistant  continues. In the future, data from applications in the field will also be  incorporated into the solution. This will enable faster and more reliable results, because the self-learning system will continuously improve. Through all of this,  TRUMPF ensures that the algorithm does not distribute the users expertise.

“Manufacturers and users alike benefit from this novel development approach. Of  course, data security remains a top priority for TRUMPF,” says Peters. TRUMPF  regularly provides users with online updates that they can download to their  machines.