Hybrid Model Driven Optimal DoE
Watch Dr. Alessandro Butté, CEO of DataHow, explain how today’s “state-of-the-art” approaches using factorial designs are no longer optimal when used alongside emerging technologies such as AI and hybrid modeling.
Speakers
-
Dr. Alessandro Butté
Founder & CEO DataHow AG
Overview
Experimental design is a key activity for planning and conducting experiments to deliver the essential process data to understand and optimize a process. DoE and factorial designs are considered “state of the art” and the industry standard – but is that still true?
Dr. Alessandro Butté, CEO of DataHow, explained how today’s “state-of-the-art” approaches using factorial designs are no longer optimal when used alongside emerging technologies such as AI and hybrid modeling.
Key topics included:
What are the limitations of factorial designs and what is “optimal”;
How a different approach is required when using hybrid models;
How to re-define optimality in the view of utility to the user;
What is the role of Bayesian optimizations and how to adapt them to hybrid models;
How we can support users to chose an optimal strategy to achieve their goals.