• 한국데이터정보과학회:학술대회논문집
• /
• 2005.04a
• /
• pp.69-82
• /
• 2005

Factorial experiments are studied in this paper. The Designs, thus, have factorial balance with respect to estimable main effects and interactions. John and Lewis (1983) considered generalized cyclic row-column designs for factorial experiments. A simple method of constructing confounded designs using the classical method of confounding for block designs is described in this paper.

• Communications for Statistical Applications and Methods
• /
• v.7 no.3
• /
• pp.759-766
• /
• 2000

In order to design fractional factorial experiments which include some debarred combinations, we should select defining contrasts so that those combinations are to be excluded. Choi(1999) presented a method of selectign defining contrasts to construct orthogonal 3-level fractional factorial experiments which exclude some debarred combinations. In this paper, we extend Choi's method to general p-level fractional factorial experiments to select defining contrasts which cold exclude some debarred combinations.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.3
• /
• pp.371-376
• /
• 2003

In this paper, we proposes a method for recognizing the facial expressions using ICA(Independent Component Analysis)-factorial representation method. Facial expression recognition consists of two stages. First, a method of Feature extraction transforms the high dimensional face space into a low dimensional feature space using PCA(Principal Component Analysis). And then, the feature vectors are extracted by using ICA-factorial representation method. The second recognition stage is performed by using the Euclidean distance measure based KNN(K-Nearest Neighbor) algorithm. We constructed the facial expression database for six basic expressions(happiness, sadness, angry, surprise, fear, dislike) and obtained a better performance than previous works.

• Journal of Korean Society for Quality Management
• /
• v.28 no.3
• /
• pp.59-67
• /
• 2000

In analysis of resolution IV fractional factorial experiments, the main effects only are analyzed, even though we can get some useful information on the confounded 2-factor interactions. In this paper, we introduce an exploiting method of the confounded structure of interactions, especially for the near minimal resolution IV 3$^{t}$ fractional factorial designs developed by Anderson and Thomas (1979). Moreover, in this paper the application way of the proposed method is also discussed by analyzing some simulated data.

• Proceedings of the Korean Statistical Society Conference
• /
• 2004.11a
• /
• pp.313-317
• /
• 2004

Confounded row-column designs for factorial experiments are studied in this paper. The Designs, thus, have factorial balance with respect to estimable main effects and interactions. John and Lewis (1983) considered generalized cycle row=column designs for factorial experiments. A simple method of constructing confounded designs using the classical method of confounding for block designs is described in this paper

• Communications for Statistical Applications and Methods
• /
• v.6 no.3
• /
• pp.695-706
• /
• 1999

When fractional factorial experiments contain some infeasible treatment combinations called debarred combinations we should construct experimental designs so that those debarred combinations are to be excluded by selecting defining contrasts appropriately. By applying Franklin(1995)'s procedure for selecting defining contrasts to Cheng and Li(1993)'s method this paper presents a method of selecting defining contrasts to construct orthogonal 3-level fractional factorial experiments which exclude some debarred combinations.

• Communications for Statistical Applications and Methods
• /
• v.15 no.4
• /
• pp.497-507
• /
• 2008

Confounding techniques have to be used repeatedly in the situations where it is necessary to perform only 2 runs under homogeneous conditions in $2^m$ factorial and fractional factorial experiment. Combinations of confounded $2^m$ factorial and fractional factorial designs enable the estimation of all main effects and all of or a part of 2 factor interaction effects. Defining contrast are used for our designs and treatment combinations of designs to be run are presented.

• Journal of Korean Society for Quality Management
• /
• v.49 no.1
• /
• pp.31-45
• /
• 2021

Purpose: A noncentral composite design method is to be developed to explore farther region for the first factorial design. A general guideline for sequential experimentation is provided. Methods: (1) A non-overlapping noncentral composite design (NNCD) is developed, in which the second factorial design shares one design point that indicates the best response value in the first factorial design. (2) Four composite designs are compared in terms of the four design evaluation criteria, which are D-, A, G, and I-optimality. (3) A follow-up design strategy is suggested based on the interaction effect, direction of improvement, number of factors. Results: (1) NNCD and model building method are presented, which is useful for exploring farther region from first factorial design block. (2) The performances of the four composite designs are compared. (3) A follow-up design strategy is suggested. Conclusion: (1) NNCD will be useful to explore farther region for the first factorial design. (2) A follow-up design strategy can be beneficial to the experimental practitioners for product and process design and improvement.

• Journal of Industrial Technology
• /
• v.21 no.B
• /
• pp.307-315
• /
• 2001

The purpose of this study was to provide the evaluation and prediction of strengths of SB latex modified concrete(LMC) using a statistical method and factorial experimental design method. The main experimental variables were as follows ; W/C ( 4 levels ; 31, 33, 35, 42%), S/a( 2 levels ; 55, 58%) and L/C(2 levels ; 5, 15%). The compressive strength and flexural strength of LMC were selected as a factor of response. The statistical method was carried out to analyze the results, together with factorial experimental design method and response surface method. The analysis showed that if L/C had been 15%, W/C appeared to be around 33% to achieve the design strength of $350kgf/cm^2$. In this case, the flexural strength and the slump came to around $68kgf/cm^2$ and 18cm, respectively. Eventhough the L/C varied, the design strength and W/C could be predictable together with slump value and flexural strength. As a result of series of experiments in this study, W/C and L/C were proved to be the main factors influencing on the compressive and flexural strength of LMC. Both of strength and slump values could be predictable from the mixing proportion of LMC.

• Journal of the Korean Statistical Society
• /
• v.22 no.2
• /
• pp.249-258
• /
• 1993

It is well known that design matrix X for any factorial design can be represented by a product $X = TX_o$ where T is replication matrix and $X_o$ is the corresponding balanced design matrix. Since $X_o$ consists of regular arrangement of 0's and 1's, we can easily find the spectral decomposition of $X_o',X_o$. Also using this we propose an efficient algorithm for computing the orthogonal projection matrix for a balanced factorial design.