摘要
在工业统计试验和其他科学试验中，常遇到因子个数多而所允许的试验次数少的情况，这时要用到超饱和设计．以前的文章仅研究了二水平因子的超饱和设计，本文对于三水平因子的超饱和设计提出了一种基于典则相关意义下的优良性准则和构造算法，并给出了试验次数为９和１８可分别安排到２６和３０个因子的三水平超饱和设计表．
In industrial statistics and other scientific experiments, there are often situations in which the number of factors is very large as compared with the number of runs being permitted. Supersaturated designs are applied to these occasions. In this pa- per, a criterion and an algorithm are given based on canonical correlations for constructing 3-level factorial supersaturated designs. With the criterion and algorithm, the 3-level su- persaturated designs with 9 runs, 26 factors and 18 runs, 30 factors are constructed and tabulated.
出处
《应用数学学报》
CSCD
北大核心
2000年第2期161-171,共11页
Acta Mathematicae Applicatae Sinica
基金
国家自然科学基金
关键词
因析试验
超饱和设计
典则相关
三水平因设子
Canonical correlation (CC), canonical correlation pattern (CCP), minimax canonical correlation criterion (mMCC-criterion), supersaturated design, 3-level factorial design
