Digital Twin Technology Reduces Waste and Enhances Efficiency in Automotive Manufacturing

Patrone and Mongiello selected Altair’s digital twin solution, implemented by Advanced Engineering (AE) Solutions, to meet their challenge. The solution involved creating a comprehensive digital twin that simulated the company’s existing sheet metal forming process. This included the machine press, sheet-metal behavior, system variables, and operating conditions. The digital twin used both simulated data and real data from accelerometers and AutoGrid sensors on the physical machine press. Altair produced the simulated data through finite element analysis (FEA) with Altair® Inspire™ Form. They also ran design-of-experiment (DoE) studies with Altair® HyperStudy® to reveal the effect of the sheet-metal properties and equipment-operation settings on the forming process performance. The team used Altair® romAI™ to create efficient reduced-order models (ROMs) that support AI and enable very fast simulation runtime. The ROMs revealed parameter variation effects on the forming process’s end-product quality. The team then used romAI to train machine learning models with simulated data. All data models were deployed through Altair® Panopticon™, a cloud-based dashboard where forming press operators could visually monitor actual sensor data against expected behavior for key performance indicators (KPIs) and deploy corrective actions throughout the sheet metal forming process.