A new series of main effects plus one plan for 2(m) factorial experiments with m=4 lambda +/- 1 and 2m runs

作者：Esmailzadeh Nabaz^*; Talebi Hooshang; Momihara Koji; Jimbo Masakazu

来源：Journal of Statistical Planning and Inference, 2011, 141(4): 1567-1574.

DOI：10.1016/j.jspi.2010.11.008

• Non-negligible effect
• Search designs
• Search linear model
• PBIBD

An MEP.1 plan for a 2(m) factorial experiment is able to search and estimate one possible nonzero interaction effect along with estimating the general mean and main effects. Construction of such a plan has been tackled by several authors. Some of these works are developed for all m, in which are focused on m = 2(h - 1) or m = 4 lambda - 1 for positive integers h > 2 and lambda >= 2. This leads to a gap for having MEP.1 plans with well-structured runs set and probably with a reasonable number of treatments for m not equal 2(h) - 1 or 4 lambda - 1. In this paper, we characterize and present a series of highly structured MEP.1 plans for all odd m = 4 lambda +/- 1,lambda >= 2, with N = 2m runs. From these designs one can obtain the MEP. I plans for even m with N = 2m + 1 treatments. The exemptions are m = 8 and 9 for which we have 1 or 2 extra runs.

• 出版日期2011-4