Shari Kraber is a consultant with Stat-Ease, with over 20 years of experience teaching engineers and technical professionals how to design and implement design of experiments. Shari draws on her process engineering experiences at 3M and Frigidaire where she implemented statistical quality tools such as statistical process control and DOE. As the Stat-Ease Workshop Manager, she now devotes her considerable energy, experience and leadership skills to workshops on DOE, teaching others how to make breakthroughs.
- Professional Engineer (PE), #26402 MN
- Bachelor of Science in Industrial Engineering, Iowa State University, 1989
- Masters of Science in Applied Statistics, Rochester Institute of Technology (RIT), 2003
This presentation introduces a simple but effective four-step DOE planning process that includes the evaluation of the power of the design to detect effects. Knowing the power prior to running an experiment forces the quality professional to consider both the amount of change in the response and the level of inherent variation in the process. This is particularly important when searching out small but very important effects on a noisy system. After completing these and the other steps in the DOE process, experiments will be far more likely to choose a test plan that fits the problem.. While laying out this path, the presentation will provide modern design options that are more effective and efficient than traditional alternatives.
The learning objectives are:
- Use a four-step DOE planning process to create an effective experiment.
- Understand how to evaluate the power of a factorial DOE prior to running the experiment. This will ensure that the chosen design can detect the effects of interest.
- Explore guidelines for choosing a test plan that meets the experimental objectives, including modern design options that maximize information while minimizing runs.