Case Study: BMW accelerates crashworthiness optimization with Altair’s AI-enhanced surrogate modeling

Crashworthiness Optimization at BMW With Ai-enhanced Surrogate Modeling

BMW, the world’s leading premium automaker, faced computationally expensive body‑in‑white crash design problems where opposing load cases, multiple KPIs (energy absorption, peak force, local displacements, weight) and discrete event timing make crashworthiness optimization complex and slow. To accelerate validation and capture expert decision‑making, BMW worked with Altair, using Altair’s integrated Machine Learning solutions within HyperWorks to augment engineering expertise.

Altair implemented an ML‑driven surrogate modeling workflow—DoE sampling, evaluation, unsupervised clustering and a classifier that enforces favorable crash kinematics during optimization—to emulate engineers’ judgments. The Altair solution simplified the optimization formulation, targeted specific crash kinematics, reduced the number of design iterations and shortened development cycles, enabling BMW to allocate computing and human resources more efficiently for high‑value simulation and validation.

BMW

Moritz Frenzel

BMW