The Designs of Blocked Fractional Factorial Split-plot Experiments in Blocks of Size Four
| Autor: | Yu-Shan Lu, 盧宥姍 |
|---|---|
| Rok vydání: | 2012 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 100 The split-plot designs are widely used in experiments around the world. Many researches in the past studied on the fractional factorial split-plot designs such as Box and Jones (1992), Haung, Chen, and Voelkel (1998), Bingham and Sitter (1999) and Bisgaard (2000). In the some situations, the experimental conditions can not remain homogeneous for all fractional factorial split-plot design, then blocking may represent a practical design option. If we only consider the blocks of size two in the split-plot design, then we can not estimate all main effects or it may violate the essence of the split-plot design. Therefore, we consider blocks of size four in the blocked fractional factorial split-plot design. We give the eight-run, 16-run, and 32-run blocked factorial split-plot designs which could estimate the most main effects and two-factor interaction effects between the whole-plot and the sub-plot factors in the full factorial design in Section 2. In Section 3, we discuss three different cases for the blocked fractional factorial split-plot design. The first one, the words of the defining relation in the subplot part only; the second, the words of the defining relation in the wholeplot part only; and the third, we consider that in both wholeplot and subplot at the same time. We present five examples, one for factorial design, and four for fractional factorial designs, in Section 4. The conclusion are given in Section 5. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|