• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.519-536
• /
• 1995

In the design of experiments for response surface analysis, sometimes it is of interest to estimate the difference of responses at two points. If differences at points close together are involved, the design that reliably estimates the slope of response surface is important. This idea was conceptualized by slope rotatability by Hader & Park (1978) and Park (1987). Until now, second order polynomial models were only studied for slope ratatability. In this paper, we will propose the necessary and sufficient conditions for slope rotatability over all directions for the thired order polynomial models in two, three and four independent variables. Also practical slope rotatable designs over all directions for two independent variables are suggested.

• 품질경영학회지
• /
• 제14권1호
• /
• pp.26-32
• /
• 1986

A class of multi-factor designs for estimating the slope of second order response surfaces is presented. For multi-factor designs the variance of the estimated slope at a point is a function of the direction of measurement of the slope and the design. If we average the variance over all possible directions, the averaged variance is only a function of the point and the design. By choice of design, it is possible to make this variance constant for all points equidistant from the design origin. This property is called "slope-rotatability over all directions", and the necessary and sufficient conditions for a design to have this property are given and proved. The class of design with this property is mainly discussed.

• Journal of the Korean Statistical Society
• /
• 제9권2호
• /
• pp.145-172
• /
• 1980

Draper and Lawrence (1965a) have given mixture designs for three factors when all the mixture components can vary on the entire factor space so that the region of interest is an equilateral triangle in two dimensions. In this paper their work is extended to the cases when the region of interest is an echelon, parallelogram, pentagon or hexagon, because of the restirctions imposed on some or all of the mixture components. The principles used in the choice of appropriate designs are those originally introduced by Box and Draper(1959). It is assumed that a response surface equation of first order is fitted, but there is a possibility of bias error due to presence of second order terms in the true model. Minimum bias designs for several cases of restricted regions of interest are illustrated.

• 대한기계학회논문집A
• /
• 제28권10호
• /
• pp.1458-1466
• /
• 2004

Deterministic optimum designs that are obtained without consideration of uncertainties could lead to unrealiable designs. Such deterministic engineering optimization tends to promote the structural system with less reliability redundancy than obtained with conventional design procedures using the factor of safety. Consequently, deterministic optimized structures will usually have higher failure probabilities than unoptimized structures. This paper proposes the reliability based design optimization technique fur apressure tank considering temperature effect. This paper presents an efficient and stable reliability based design optimization method by using the advanced first order second moment method, which evaluates a probabilistic constraint for more accuracy. In addition, the response surface method is utilized to approximate the performance functions describing the system characteristics in the reliability based design optimization procedure.

• Journal of the Korean Statistical Society
• /
• 제11권1호
• /
• pp.36-44
• /
• 1982

A new design concept, called axis-slope-rotatability, is presented for the design of experiments with mixtures. This is an analogue of the Box-Hunter (1957) rotatability for second order response surface designs. By choice of design, it is possible to make the variance of the estimated slopes along the component axes constant for all axial points equidistant from the center point of the factor space. This property is called axis-slope-rotatability for mixture experiments. When the Scheffe's second degree polynomial is used, it is shown that some symmetry conditions are sufficient for axis-slope-rotatability. Several designs having this property are illustrated.

• 대한안전경영과학회지
• /
• 제15권4호
• /
• pp.123-129
• /
• 2013

Among various manual materials handling tasks, pushing/pulling was known to be one of the risk factors for the low back and shoulder musculoskeletal disorders (MSDs). This study was conducted to find out an optimal solution set of the handle height and distance for 4-wheel cart with two vertical handles. Ten male college students participated in the pushing force measurement experiment. The face-centered cube design, one of the central composite designs, was applied for the experiment, and the isometric voluntary pushing force was measured in 9 treatment conditions. The second order response surface model was predicted by using the pushing strength as a response variable, and the handle height and distance as independent factors. According to the 2nd order response model, the handle height and distance showed nonlinear relationship with the isometric pushing strength. To maximize the 2nd order response model (pushing force), the handle height and distance were optimized. The optimal handle height was 'xyphoid process height - stature', and the optimal handle distance was '$1.25{\times}shoulder$ width'. When calculated using the anthropometric data of the subjects of this study, the optimal handle height was $115.4{\pm}3.4$ cm, slightly higher than the elbow height, and the handle distance was $52.9{\pm}2.3$ cm.

• Communications for Statistical Applications and Methods
• /
• 제9권3호
• /
• pp.595-605
• /
• 2002

The central composite design is widely used for estimating second order response surfaces. This type of design is composed of $2^k$ factorial points, axial points and center points. In this paper, we suggest a version of central composite design where the positions of the axial points are indicated by two numbers, and study properties of this design. We obtain the variances and covariances of the estimators of the regression coefficients. Conditions are obtained for this design to be orthogonal and rotatable. This design is compared with other designs on the basis of efficiency.

• Journal of the Korean Statistical Society
• /
• 제27권1호
• /
• pp.133-152
• /
• 1998

The central composite design is widely used in the response surface analysis, because it can fit the second order model with small experimental points. In practice, the experimental data are not always obtained on all the points. When there are missing observations, many problems due to the missing cells can occur. In this paper, the influence of a missing cell on the central composite design is discussed. First, the influences of a missing cell on the variances of estimated regression coefficents are compared as $\alpha$ varies. Second, how the average predition variance is affected by a missing sell is discussed. And the influence on rotatability is investigated. Third, the influence of a missing cell on optimality, especially on D-optimality and A-optimality, is examined.

• 응용통계연구
• /
• 제20권2호
• /
• pp.267-280
• /
• 2007

혼합물 성분들의 비율의 상한과 하한에 대한제한조건이 부과된 제한된 혼합물 실험 공간 R에서의 혼합물 실험을 위한 최적 설계를 찾는 데에 D-, G-, V- 최적기준 등과 같은 다양한 최적 설계 기준이 사용된다. 각각의 실험 설계는 선택된 최적 기준에 대해서는 최적이지만, 제한된 혼합물 실험 공간에서의 예측력에 대해서는 만족스럽지 못하다는 것은 잘 알려진 사실이다. (Vining 등, 1993; Khuri 등, 1999). 우리의 관심사는 2차 혼합물 반응표면모형을 가정한 경우에 제한된 혼합물 공간에서의 효율적인 실험 설계를 찾는 것이다. 이 논문에서는 꼭지점, 선중심점, 면중심점, 중앙점과 내부점으로 구성된 확장된 후보 실험점 그룹을 구성한 다음에, D-최적기준, G-최적기준, V-최적기준과 실험점들 간의 거리에 근거한 U-최적기준에 강건한 실험 설계를 제안한다. Khuri 등(1999)에서 분석된 비료 혼합물 실험과Vining과 Cornell(1993)이 분석한 조명탄 혼합물 실험의 사례에서 강건한 실험설계들과 두 논문에서 추천된 실험 설계들에 대한 예측치의 표준화된 분산의 분위수의 그림(SVPQP)을 비교한 결과 강건한 설계가 상대적으로 우월함이 판명되었다.

• 응용통계연구
• /
• 제25권1호
• /
• pp.215-223
• /
• 2012

반응표면 방법론은 어떤 공정을 개선하거나 최적화하는데 이용되는 아주 유용한 통계적방법이다. 이러한 최적조건을 추정하기 위하여 최적조건이 있으리라 예상되는 실험구역을 탐색하여 실험을 실시한다. 그런데 이 실험구역은 실험의 환경의 제약 그리고 연구자의 선택 등으로 그 모습이 다양하게 달라질 수 있다. 반응표면 설계는 실험구역의 모양에 따라 보통 둥그런 모양의 "구형설계"와 육면체 모양의 "입방형설계"로 구분한다. 구형설계는 회전성을 만족하거나 회전성에 상당히 근접하는 "유사회전성"을 갖는 특징이 있다. 반응표면 설계에서 가장 많이 사용되는 중심합성설계는 실험구역이 구형인 5수준 실험설계이다. 이 때, 축점의 ${\alpha}$값을 ${\alpha}=\sqrt{k}$ 대신 ${\alpha}=1$로 조정하면 5-수준이 아닌 3-수준 입방형 중심합성설계를 얻을 수 있다. 그러나 입방형 중심합성설계는 실험구역이 구형이 아니므로 회전성을 만족하지 못하는 문제가 있다. 이러한 이유로, 변수들의 수준 수를 3으로 제한하면서 실험구역은 구형인 실험설계가 필요할 때가 많다. 이에 대한 대표적 실험설계가 바로 박스-벤켄 실험설계이다. 이 실험설계는 구형의 실험구역으로 회전성을 만족하나 실험구역의 크기가 변수의 개수가 증가해도 제자리 수준으로 좁은 특징이 있다. 현실적으로 실험구역의 가상 자리 부분에 대한 예측에 관심이 있을 경우 변수의 개수가 많아지면 이에 비례하여 실험구역이 커지는 실험설계가 바람직하다. 본 논문은 3-수준 입방형설계에 비하여 실험구역이 유달리 좁은 박스-벤켄 실험설계를 보완하여 구형설계를 만족하면서도 다른 한편으로는 변수 수에 따라 실험반경이 커지는 3-수준 구형 반응표면 설계를 소개하고자 한다. 이 방법을 기존의 실험설계들과 비교한 결과 변수수가 비교적 작을 경우 실험횟수 등을 고려하여 응용가치가 있음을 확인하였다.