This course explores how experimental design can be applied for product and process optimization of a response variable such as yield, performance, cycle time, or cost reduction. Using Response Surface Methods, second order models are created to relate process inputs to process outputs allowing settings for the inputs to be determined that will achieve optimal performance for the outputs.
Process optimization methods are given for single as well as multiple outputs considered simultaneously using desirability functions and includes a formal method for experimental confirmation.
Participants have an opportunity to practice these methods by designing, running, and analyzing a series of experiments via unique computer simulations, providing an experience as close as possible to actually running industrial experiments. A final course project lets teams compete to determine which team can find the best optimal solution.
Who Should Attend
This course is designed for engineers, quality professionals, researchers, and managers who need to understand and use DOE for process optimization.
Through training, participants will:
- Understand response surface strategy for process optimization
- Be able to optimize one as well as multiple response variables simultaneously
- Know how to perform what-if scenarios to examine practical solutions
- Be able to conduct confirmation runs using prediction intervals to validate a DOE
Introduction to Response Surface Methods
- Goals and Strategy of RSM
- Case Study
- Experimental Designs for Fitting First Order Models
- Path of Steepest Ascent
- Experimental Designs for Fitting Second Order Models
- Central Composite Designs
- Box-Behnken Designs
- Rotatability and Orthogonality
- Blocking an RSM Design
- Group RSM Exercise 1
- Group RSM Exercise 2
Multiple Response Optimization
- Desirability Function
- Multiple Response Optimization
- Confirmation Run Verification
- Final Group Project Competition
Basic DOE or the equivalent
Instructor-led class training, with opportunities to practice learned skills using prepared data, live demonstrations, and data collected real time in class using computer simulations
Minitab or JMP Statistical Software