Foldover design is a commonly used technique that is obtained by adding another fraction to an initial fractional factorial design. Because such added designs require the same run size as the initial design, semifoldover designs that consist of half of a foldover fraction have been investigated in the recent literature for regular two-level fractional factorial designs. As the initial design and the follow-up semifoldover design are usually conducted at different stages, it is important to consider a block factor reflecting this effect. This paper studies the impact of the block factor on the semifoldover of 2k-p designs. We first propose a method for obtaining the equivalent semifoldover plans. Then the properties and the structures of blocked semifoldover designs are explored. The optimal blocked semifoldover designs for 16 and 32 runs are obtained and tabulated for practical use.
|Original language||English (US)|
|Number of pages||8|
|Journal||Journal of Statistical Planning and Inference|
|State||Published - Jan 2014|
Bibliographical noteFunding Information:
We would like to express our sincere thanks to the editor and two anonymous reviewers for their valuable comments and suggestions. Yang's research was supported by the Summer Research Grant of DePaul University . Li's research was supported by the Research And Teaching Supplements System, Carlson School of Management, University of Minnesota .
- Block factor
- Optimal plans