 |
Robust Design for Six Sigma
Douglas C. Montgomery, Ph.D.
G. Geoffrey Vining, Ph.D.
Thursday - Friday
February 24 - 25, 2005
|
| |
|
| Text: Response Surface Methodology, 2nd edition, R. H. Myers and D. C Montgomery, (John Wiley, NY, 2002). All participants will receive a copy of this textbook plus accompanying workbook notes prepared by Dr. Montgomery. |
| |
|
| Design for Six Sigma (DFSS) can have a significant impact on the product realization process, leading to products that are easier to manufacture, have lower life-cycle costs, and result in higher levels of customer satisfaction. Many DFSS problems involve variables that are difficult to control when the product/process is in full-scale operation or used by customers. These variables may be environmental in nature, or they may be components, material properties, or other factors that transmit unwanted variability into the final product. The objective in these situations is to make the product or process robust to this unwanted variability. Statistical experimental design, modeling building, and optimization techniques are the modern, easily implemented basis for solving these problems. These modern approaches require less effort and fewer resources than earlier methods for achieving product/process robustness, and they lead to more efficient solutions to the problem. This course provides a thorough treatment of these techniques and shows how that can be easily applied in the DFSS environment. |
|
|
|
Course Outline
|
 |
The robust parameter design and process robustness problem |
|
Controllable variables and noise variables including early approaches, Taguchi's definition and solution of the problem, signal-to-noise ratios and crossed array designs based on orthogonal arrays, and problems with this approach |
|
The modern approach: a response model solution |
|
Building a response model containing both controllable and noise variables |
|
Finding the transmitted variability |
|
System optimization including minimizing variability around a target, maximization or minimization of the response while minimizing variability, handling multiple responses, and multiple design objectives |
|
Experimental designs for robust design and process robustness studies including modified response surface designs, mixed resolution designs, computer generated designs, and guidelines for choice of designs |
|
Special applications including computer models for product/process design, finite element models, and other types of simulators |
|
| This course contains material necessary for the ASQ Certified Quality Engineer, Certified Quality Engineer-in-Training, and Certified Quality Auditor examinations. CEU credit will be added upon completion of this course |
|
