Webinars

Taguchi methods have been widely used for robust design and optimization, particularly to improve product quality by reducing sensitivity to noise factors. However, modern engineering simulations often involve complex input-output relationships, multiple forms of uncertainties, and numerous inputs, which can benefit from modern uncertainty quantification (UQ) techniques.

UQ uses space-filling design of experiments (DOE) and advanced statistical and machine learning models to accelerate simulations and perform uncertainty analysis that would otherwise be computationally too expensive. Unlike traditional factorial designs in Taguchi methods, UQ’s space-filling designs more efficiently sample the design space and better predict complex, nonlinear simulation behavior.

As a modern extension of Taguchi's robust analysis, UQ supports:

• Sensitivity Analysis to identify key contributors to uncertainty.
• Uncertainty Propagation to evaluate how input uncertainties affect outputs.
• Stochastic Optimization to optimize design factors under uncertainty.
• Statistical Calibration to improve simulation accuracy under parameter and model form uncertainty.

Join us for this webinar, where SmartUQ’s Principal Application Engineer, Gavin Jones, will showcase SmartUQ’s tools for integrating Taguchi methods with UQ and explore capabilities in space-filling designs, machine learning, optimization under uncertainty, and simulation model calibration.