Menu
Log in
 

International Statistical Engineering Association

A case study of a mixed-level OMARS design

  • 20 Sep 2023
  • 10:00 AM - 11:00 AM
  • Online


Maria Lanzerath, statistician at W. L. Gore & Associates, will give the next ISEA webinar on design of experiments and showcases how these are used in her work.

OMARS (orthogonal minimally aliased response surface) designs represent a new class of highly efficient designed experiments with a true potential to combine a screening experiment and a response-surface experiment into just one. This leads to significant savings in time and resources, which can be particularly critical for costly experiments. In the case study that is presented, an OMARS design was conducted for the development and validation of a chemical product, a new resin. The model building and design finding, setup, model analysis for several responses, and the results will be explored. The design had eight factors in total, with five of them executed on three levels, and three on two levels, therefore a mixed-level design. In addition, covariates were considered in the analysis.

Maria Lanzerath leads a global team of statisticians at W. L. Gore & Associates, working with scientists and engineers to ensure that statistics is embedded throughout product development and the manufacturing process. She has been an industrial statistician with the company since 1997, providing expertise in designed experiments, statistical process control and data-driven decision making. Lanzerath led the development of a companywide DOE training class for hundreds of engineers and scientists in the US, Germany, Japan, and China. Before joining Gore, Lanzerath studied statistics at the Ludwig-Maximilians University in Munich and worked for three years as a data manager and SAS programmer for a Japanese pharmaceutical company. Lanzerath innovated the usage of DOE at Gore, moving from traditional fractional factorial designs to optimal designs.

View Webinar Slides

View Webinar Recording

Copyright 2020 International Statistical Engineering Association

Join us on

 

Powered by Wild Apricot Membership Software