Robust Design Principles and Applications

no comments
Aug 17 - Aug 21

Variation in the performance of an engineered system can come from sources including part-to-part variation, uncontrolled noise parameters, our own assembly processes, or from the design itself. According to Dr. Genichi Taguchi, robust design is the concept of reducing variation in the performance of a system without addressing the causes of variation. In short, using principles of robust design, we make the system insensitive to variation in low-level design parameters.

SigmaPro’s Robust Design Principles and Applications course has been specifically developed to provide industrial Black Belts, Master Black Belts, product/process development engineers, and other technologists with a concise and applied treatment of the latest methods in robust design. Technical tools to accomplish robust design include requirements definition, qualitative decomposition, transfer function modeling for one or more responses, designed experiments using inner/outer arrays, probabilistic design, statistical tolerancing, and much more. To facilitate learning and increase knowledge retention, numerous hands-on examples and exercises will be covered in detail. A capstone exercise enables all participants to demonstrate their mastery of the robust design process from start to finish.

Candidate Qualifications

- Candidates for this course are typically industrial personnel with previous Green Belt, Black Belt, or Master Black Belt certification as well as product design & process development engineers.
- The majority of previous candidates have come from aerospace, automotive, chemical, pharmaceutical, medical device, manufacturing, and electronics sectors.
- A training certificate will be provided which may be used for recertification credits.

Participants Will Learn

- How to develop detailed design requirements from higher level system requirements.
- How to select the best design concept based on customer requirements.
- How to use qualitative decomposition to identify key design nodes with respect to risk, identify all input and output variables, and determine the reduced set of key inputs and outputs for further optimization.
- How to examine and optimize a system using Axiomatic Design.
- How to analyze one or more responses and build effective prediction models using transfer function modeling techniques.
- How to employ effective inner/outer array structures with any designed experiment to accomplish screening, prediction, or optimization.
- How to use nonlinear optimization techniques to find optimal values for design parameters for one or more response functions.
- How to set optimal statistical tolerances based on cost and system requirements.
- When and how to use concepts such as loss functions, signal to noise ratios, parameter design, the energy function, and the Mahalanobis distance.
- How to put the above-mentioned tools to use as an effective design strategy during the Capstone exercise.