Altair, a global leader in computational intelligence, will show at HANNOVER MESSE 2025 how its cutting-edge artificial intelligence (AI) and simulation capabilities are transforming industries. Altair will demonstrate how its technology is helping companies innovate faster, optimize performance and achieve sustainable success through AI-powered engineering, high-performance computing (HPC), knowledge graphs, digital twins and more.

“AI and machine learning are critical to achieving the quartet of speed, performance, quality and cost efficiency that today's organizations seek. Altair's AI-powered engineering solutions provide deeper insights, accelerate design and improve decision-making to create a safer, more connected and sustainable future,” said Ravi Kunju, Chief Product and Strategy Officer at Altair.

Many control systems – especially those that support autonomous driving – require complex, realistic system models. Physical simulations, together with AI-supported modeling at the system level, can provide realistic synthetic data to capture the interactions between the machine and the material, the terrain and the subsurface.

At HANNOVER MESSE 2025, Altair will be presenting impressive customer examples, including an application at CNH Industrial. Here, Altair's AI-powered simulation tools enabled a powerful digital twin that accelerates and optimizes product development through rapid prediction. “Altair's innovative solutions for virtual product development have enabled us to create a digital twin of our machine that helps us gain a better understanding of machine behavior and reduce time to market,” said Giuseppe Gullo, FEA Design Analysis Engineer, CNH Industrial. “The AI algorithm maps complex effects and accelerates the delivery of system responses for an efficient overall view. This has allowed us to streamline validation, reduce computational costs and optimize our development process.”

The technology behind it: an AI module for the dynamic generation of reduced-order models. Reduced-order models (ROMs) are useful for incorporating detailed 3D simulations into a computationally more efficient 1D environment for system-level investigations. Simulation tools such as Altair EDEM or Altair CFD enable detailed investigations of time-varying and nonlinear systems. Due to the long runtimes of the simulations, the analysis usually focuses on one component or one subsystem. However, in the case of a complete system simulation, it is often sufficient to reduce the behavior of the components to their interaction with the complete system in order to shorten the runtime of the solver and still obtain sufficiently accurate results.

By using Altair's romAI artificial intelligence tool, 3D simulations can be used as training data to create dynamic ROMs. Fewer 3D simulation runs are required because this approach requires less training data than conventional data-driven methods. Altair says that romAI can work with any solver and can provide highly accurate results when operating within the training domain, and is said to be stable and useful even when extrapolating outside the domain. The same machine learning technique can also be used for system identification when starting from test data.

Video