Reducing Design Cycles with Statistics and Machine Learning

Reducing Design Cycles with Statistics and Machine Learning

Machine learning (ML) models trained on simulation or manufacturing data can help accelerate analyses crucial to meeting quality requirements. This presentation will introduce design of experiment (DOE) and data sampling tools for collecting or identifying the data needed for ML model training. The use of ML models for uncertainty propagation and sensitivity analysis to quantify manufacturing uncertainty, better understand margins, and identify the biggest drivers of uncertainty will also be discussed. Points will be illustrated with several examples from SmartUQ customers.

Speakers