FACE TO FACE
TRUMPF ’ s digital run-in assistant in the TruMatic 5000 punch laser machine .
TRUMPF ’ s digital run-in assistant automatically determines which parts require manual set-up . For example , the solution calculates a part ’ s bending angle during removal .
intelligence can be used in manufacturing as well as the benefits that it can bring to manufacturers of all sizes .
“ There are numerous opportunities for AI in industrial production . There are so many manufacturers who are not aware of what it can do in production and logistics to optimise processes . Simulations can be ten thousand times faster if you use AI to build machines , for example . It is really important to bring this new technology into more companies ,” explained Professor Sepp Hochreiter , University of Linz .
Stephan Mayer illustrated how TRUMPF is using AI to improve its customers ’ processes ( more stable , rapid and efficient ) and increase uptime of its customers ’ machines . It also uses AI to increase the efficiency of its own internal processes such as order administration , issue resolutions or R & D .
“ AI can recognise repeating patterns and structures from Big Data so if you have data such as pictures , sound , text or video for example , AI helps you to recognise patterns out of the data and make better decisions for better statistical outcomes . So , we are therefore looking for customer
TRUMPF demonstrated its ‘ Runability Guide ’ AI solution at INTECH .
Hagen Zimer ( CEO , TRUMPF Laser Technology ).
Instead of improving the parts using trial and error , the software automatically shows where the greatest potential for optimisation lies
issues , challenges and problems where AI can help . If someone is manually repeating the same operation several times per day , then AI can help to automate this process . Preventive maintenance is an example of this , on the customer side . Applying AI to analyse temperature rises in a machine , for example . AI can recognise issues with welding seams from an image so that the welding process can be adjusted . There are hundreds of applications i . e . spare part ordering through the SAP etc .,” Mayer highlighted .
Artificial intelligence , he told the audience , can address the lack of skilled workers in the industry by automating the entire sheet metal process and tackling any automation failures , delays or bottlenecks ( such as removing cut
parts from a sheet metal skeleton , which needs to be done by hand ).
“ At our INTECH exhibition , we demonstrate how AI algorithms are taught which shapes ( out of hundreds or thousands ) can be automatically sorted and which shapes have issues with sorting . It will then alert the operator which shapes should be cut at night ( lights-out solutions ) or during the day ( when operators are around to manually remove cut parts from skeletons ). So , AI can help to increase the robustness of automation ,” outlined Mayer .
“ In its simplest form , AI is a concept to artificially process date to derive some benefit from it . So , when do you apply AI ? Simply when the analytics are failing . In a laser welding application , for example , you have a complex weld pool and you need to be 100 % sure that there are no leaks . If you can observe the process in closed loop control with the laser , and analyse the data to improve the process , you can add benefit / value to the application and to the customer . Battery manufacturing has a high scrap rate of around 50 %. AI , in our eyes , is an enabler to drive innovation and to bring down the scrap level much faster by process control at our customer ’ s site ,” outlined Hagen Zimer ( CEO , TRUMPF Laser Technology ).
TRUMPF ’ s culture of innovation is characterised by high-tech production , so it produces huge laser systems used to manufacture powerful microchips .
“ In our internal production , where complexity is high i . e . when building semiconductor lasers , you have contamination ( even in a clean room ). That can be damaging so we are already applying AI for image recognition and high-speed visual inspection for potential contamination . This brings the production cost down and increases our yield rates , boosting
ISMR May 2024 | ismr . net | 33