• 응용통계연구
• /
• 제5권2호
• /
• pp.169-179
• /
• 1992

블록 계획에서 요인의 효과를 독립적으로 추정 할 수 있도록 그 효과를 직교되게 구성 했을 때 그 계획을 직교 요인 계획이라 한다. 빈 칸이 없는 요인 행-열 계획의 성분계획들이 직 교된 요인의 성질을 지니면 그 행-열 계획 또한 직교된 요인의 성질이 유지 된다는 사실이 입증되었다. 이러한 이유로 빈 칸이 없는 직교요인 행-열 계획은 직교된 성질을 가지는 일 차원 성분 계획들을 단순히 결합함으로써 만들 수 있다. 그러나, 이러한 관계는 빈 칸이 있 는 요인 행-열 계획에서는 성립되지 않는다. 빈칸이 있는 요인 행-열 계획의 연구는 근래에 와서 그 필요성으로 활발히 진행되고 있다. 본 논문에서는 직교된 요인의 성질을 갖는 빈 칸이 있는 요인 행-열 계획의 일부를 조정 직교된 행-열 계획에서 행과 처리를 바꿈으로 설계해 보았다.

• 응용통계연구
• /
• 제11권2호
• /
• pp.303-315
• /
• 1998

요인실험에서 어떤 처리조합은 조작상 또는 경제적인 이유로 실험할 수 없는 경우가 있다. 이러한 실험불가능한 처리조합을 포함하고 있는 일부실시법은 불균형적인 처리조합을 구성하게 되어 어떤 요인효과에 대해 추론할 수 없게 된다. 본 논문은 실험불가능한 처리조합을 포함하지 않는 3$^{n-p}$ 일부실시법을 실행하기 위한 정의 대비의 선택 방법을 논한다한다

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2009년도 춘계학술대회
• /
• pp.423-427
• /
• 2009

This study proposes the generation mechanism of various design matrix using generalized linear model for design of experiment. Design generation method of GLM analysis, factorial design(FD) with center points, ANOVA design with lack-of-fit test, and response surface design are introduced. In central composite(CC) design, orthogonal blocking and fractional factorial design(FFD) are presented. We compare the design of Box-Benhken(BB) and face-centred central compsite design.

• Journal of Mechanical Science and Technology
• /
• 제14권9호
• /
• pp.947-955
• /
• 2000

This paper proposes an optimal design scheme to improve the muffler's capacity of noise reduction of the exhaust system by combining the Taguchi method and a fractional factorial design. As a measuring tool for the performance of a muffler, the performance prediction software which is developed by Oh, Lee and Lee (1996) is used. In the first stage of a design, the length and radius of each component of the current muffler system are selected as control factors. Then, the $L_{18}$ table of orthogonal arrays is adopted to extract the effective main factors. In the second stage, the fractional factorial design is adopted to take interactions into consideration, which the $L_{18}$ table of orthogonal arrays can not consider. For an optimal design, the $L_{27}$ table of orthogonal arrays with main and interaction effects is proposed and the noise factors such as temperature, background noise and humidity are analyzed for more efficient design simultaneously.

• KSBB Journal
• /
• 제17권3호
• /
• pp.228-234
• /
• 2002

Factorial design model을 이용하여 A. xylinum KJ1의 BC생산을 위한 최적 배양조건을 결정하였다. 요인 분석을 위한 실험 계획법으로는 부분요인 분석을 통한 factorial model을 이용하였으며, 주요 실험인자인 탄소원 농도, 교반속도(rpm), 산소분압, CSL의 농도의 네 가지 factor의 영향에 의한 BC생산량의 변화를 측정하였다. SAS 프로그램을 이용하여 전체 24개의 실험계획에서 각각의 factor의 조합에 의한 BC 생산성에 관한 결과 및 예측값을 비교한 결과 BC 생산량 면에서는 상관계수($R^2$)가 0.91으로 이었고, 수율면에서는 상관계수($R^2$)는 0.81이다. 최적 BC 생산을 위한 각 factor들이 탄소원의 농도 4%, 교반속도 460 rpm, 산소분압 0.28 atm, CSL 농도 6%일 때, 이때의 BC 생산량은 11.67 g/L로 예측되었다. 그리고 BC 생산 수율면에서 최적 배양조건이 탄소원 농도 4%, 교반속도 564 rpm, 산소분압 0.21 atm, CSL 농도 2%일 때 최적의 수율 0.42 g/g를 얻을 수 있을 것으로 예측되었다. 결정된 최적 조건에서의 실증 실험 결과 11.47 g/L의 BC 생산량을 얻을 수 있었으며, 이는 fermentor상에서의 기본실험에서 얻은 4.83 g/L보다 2.4배 이상 향상된 결과이다.

• 산업기술연구
• /
• 제21권B호
• /
• 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.

• 한국자동차공학회논문집
• /
• 제7권5호
• /
• pp.121-129
• /
• 1999

Recently, the regulations from the govemment and the concems of the people give rise to the interest in exhaust noise of passenger car as much as other vehicles. The exact analysis of various mufflers is needed to reduce the level of exhaust noise. In this paper, we propose a design to improve the mufflers capacity by reducing noise of exhaust system combining Taguchi method and fractional factorial design. In order to measure the performance of a muffler, the performance prediction software which is developed by the Dept. of Automotive Engineering at Hanyang University is used. From the current muffler system we select control factors such as lenght and radius of each component that are thought to be effective on capacity of muffler. Factors are arranged using L18, L27 table of orthogonal array and the fractional factorial design for analysis. We find some significant interaction effects using 1/3 fractional factorial design and accomplish the reduction of noise from the muffler.

• 품질경영학회지
• /
• 제44권4호
• /
• pp.799-811
• /
• 2016

Purpose: The conjoint analyst in marketing are interested in detecting whether there exist synergy or antagonistic effects between two attributes. In the cases where attributes have two or three levels, we research on the design of survey questionnaire to estimate all the main effect and as many two factor interaction effects as possible. Methods: We consider the balanced incomplete block (BIB) mixed level factorial design $2^f{\times}3^g$ or fractional factorial design. To reduce the number of questions in a questionnaire, we propose the balanced incomplete block mixed level design with minimum aberration which is generated by implementing proc factex in SAS. Also, we propose using two or three level BIB factorial design instead of mixed level designs by transforming three level attributes into two attributes of two levels and two level attribute into three level attribute by using dummy level technique. Results: We propose three methods for designing survey questionnaire where the block and design generators are found with practical number of questions in a questionnaire. By analyzing all the respondents survey data generated by the simulation study, we find the proper model and do the concepts optimization. Conclusion: The proposed methods of designing survey questionnaires seem to perform well in the sense that the proper model, and then the optimal concept is found in a case study where all the respondents survey data are generated by the simulation study.

• 품질경영학회지
• /
• 제45권4호
• /
• pp.889-902
• /
• 2017

Purpose: The orthogonality and trace optimal properties are desirable for constructing designs of experiments. This article focuses on the determination of the sizes of experiments for the balanced trace optimal resolution-V fractional factorial designs for 2-level factorial designs, which have near-orthogonal properties. Methods: In this paper, first we introduce the trace optimal $2^t$ fractional factorial designs for $4{\leq}t{\leq}7$, by exploiting the partially balanced array for various cases of experimental sizes. Moreover some orthogonality criteria are also suggested with which the degree of the orthogonality of the designs can be evaluated. And we appraise the orthogonal properties of the introduced designs from various aspects. Results: We evaluate the orthogonal properties for the various experimental sizes of the balanced trace optimal resolution-V fractional factorial designs of the 2-level factorials in which each factor has two levels. And the near-orthogonal 2-level balanced trace optimal resolution-V fractional factorial designs are suggested, which have adequate sizes of experiments. Conclusion: We can construct the trace optimal $2^t$ fractional factorial designs for $4{\leq}t{\leq}7$ by exploiting the results suggested in this paper, which have near-orthogonal property and appropriate experimental sizes. The suggested designs can be employed usefully especially when we intend to analyze both the main effects and two factor interactions of the 2-level factorial experiments.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2013년도 춘계학술대회 논문집
• /
• pp.927-928
• /
• 2013