Sequential Experimentation Schemes for Resolution III, Robust and Mixedlevel Designs
General augmentation techniques such as foldover and semifold have been a common practice in industrial experimentation for many years. Even though these techniques are extremely effective in maintaining balance and orthogonality, they possess serious disadvantages such as the inability to decouple...
| Other Authors: | |
|---|---|
| Format: | Others |
| Language: | English English |
| Published: |
Florida State University
|
| Subjects: | |
| Online Access: | http://purl.flvc.org/fsu/fd/FSU_migr_etd-1852 |
| Summary: | General augmentation techniques such as foldover and semifold have been a common practice in industrial experimentation for many years. Even though these techniques are extremely effective in maintaining balance and orthogonality, they possess serious disadvantages such as the inability to decouple specific terms and a high level of inefficiency. This dissertation aims for a sequential experimentation approach capable of improving the drawbacks of the general methods while maintaining some of its benefits. Chapter 3 begins with proposing an algorithm for sequential augmentation of fractional factorial designs resolution III. The proposed algorithm is compared with its competitors, semifold and foldover using simulated data under 3 noise level conditions. Advantages, limitations, and potential benefits of the new method are provided. Chapter 4 explores new possibilities for augmentation of efficient mixed-level designs (EAs). Current augmentation methods for mixed-level designs include only the optimal foldover plans developed by Guo (2006). Semifold plans for several mixed-level designs are developed by selecting half of the treatment combinations of the foldover fraction using the general balance metric criterion and an exhaustive search approach. Chapter 5 complements this research by providing a methodology for sequential augmentation of mixed resolution robust designs. The work presented here extends the current limits of sequential experimentation for resolution III, mixed-level and robust designs and provides a viable alternative for the experimenter in situations in which financial restrictions do not allow the implementation of a general method. === A Dissertation Submitted to the Department of Industrial and Manufacturing Engineering in Partial Fulfillment of the Requirements for the Degree of Doctor of
Philosophy. === Summer Semester, 2008. === July 11, 2008. === Semifold, Foldover, Level, Mixed, Robust, Resolution, Sequential, Experimental Design === Includes bibliographical references. === James R. Simpson, Professor Directing Dissertation; Fred Huffer, Outside Committee Member; Joseph J. Pignatiello, Jr., Committee Member; Marcus Perry, Committee Member. |
|---|