• Proceedings of the Korean Society for Quality Management Conference
• /
• 1998.11a
• /
• pp.580-593
• /
• 1998

The small composite design for second order response surface might be appropriate when experimentation is expensive, difficult, or time-consuming, especially when an independent estimate of experimental error is available. It is important that the small composite designs for response surface analysis would be rotatable and slope-rotatable. Therefore the small composite designs are studied from the viewpoint of rotatability and slope-rotatability, and the optimal values of a(the distance of axial points from the center) are investigated as k(the number of independent variables) and $n_0$(the number of center points) are changed.

• Communications for Statistical Applications and Methods
• /
• v.28 no.6
• /
• pp.595-610
• /
• 2021

Recently a few articles have derived robust first-order rotatable and D-optimal designs for the lifetime response having distributions gamma, lognormal, Weibull, exponential assuming errors that are correlated with different correlation structures such as autocorrelated, intra-class, inter-class, tri-diagonal, compound symmetry. Practically, a first-order model is an adequate approximation to the true surface in a small region of the explanatory variables. A second-order model is always appropriate for an unknown region, or if there is any curvature in the system. The current article aims to extend the ideas of these articles for second-order models. Invariant (free of the above four distributions) robust (free of correlation parameter values) second-order rotatable designs have been derived for the intra-class and inter-class correlated error structures. Second-order rotatability conditions have been derived herein assuming the response follows non-normal distribution (any one of the above four distributions) and errors have a general correlated error structure. These conditions are further simplified under intra-class and inter-class correlated error structures, and second-order rotatable designs are developed under these two structures for the response having anyone of the above four distributions. It is derived herein that robust second-order rotatable designs depend on the respective error variance covariance structure but they are independent of the correlation parameter values, as well as the considered four response lifetime distributions.

• The Korean Journal of Applied Statistics
• /
• v.25 no.1
• /
• pp.215-223
• /
• 2012

Response surface methodology(RSM) is a very useful statistical techniques for improving and optimizing the product process. By this reason, RSM has been utilized extensively in the industrial world, particularly in the circumstances where several product variables potentially influence some quality characteristic of the product. In order to estimate the optimal condition of product variables, an experiment is being conducted defining appropriate experimental region. However, this experimental region can vary with the experimental circumstances and choice of a researcher. Response surface designs can be classified, according to the shape of the experimental region, into spherical and cuboidal. In the spherical case, the design is either rotatable or very near-rotatable. The central composite design(CCD)s widely used in RSM is an example of 5-level and spherical design. The cuboidal CCDs(CCDs with ${\alpha}=1$) is appropriate when an experimental region is cuboidal but this design dose not satisfy the rotatability as it is not spherical. Practically, a 3-level spherical design is often required in the industrial world where various level of experiments are not available. Box-Behnken design(BBD)s are a most popular 3-level spherical designs for fitting second-order response surfaces and satisfy the rotatability but the experimental region does not vary with the number of variables. The new experimental design with expanded experimental region can be considered if the predicting response at the extremes are interested. This paper proposes a new 3-level spherical RSM which are constructed to expand the experimental region together with number of product variables.

• The Korean Journal of Applied Statistics
• /
• v.6 no.1
• /
• pp.105-123
• /
• 1993

반응표면의 기울기를 추정하기 위한 실험계획법이 가질 수 있는 바람직한 성질로, Hader와 Park(1978)이 제시한 "축 방향에 걸친 기울기 회전성"과, Park(1987)이 제시한 "모든 방향에 걸친 기울기 회전성"이 있다. 또한 주어진 임의의 실험계획에 대하여 축 방향에 걸친 기울기 회전성의 정도를 수치로 나타낼 수 있는 측도(measure)가 Park과 Kim(1992)에 의해 제시된 바 있다. 본 논문에서는 반응표면 실험계획법이 가지고 있는 모든 방향에 걸친 기울기 회전성의 정도를 알 수 있게 해 주는 측도를 개발하였다. 또한 이 측도를 여러 종류의 계획들에 적용하여 결과를 관찰하였다. 이 측도의 장점 중의 하나는 어떠한 계획에도 적용이 가능하다는 점이다. 계획에도 적용이 가능하다는 점이다.

• Journal of the Korean Statistical Society
• /
• v.8 no.1
• /
• pp.37-64
• /
• 1979

Criteria and designs are developed for estimating derivatives of P-variable second order polynomial response surfaces. The basic criterion used is mean square error of the estimated derivative, averaged over all directions and then averaged over a region of interest. A new design concept called slope-rotatability is introduced. A simplex optimization program is used to find minimum mena square error designs for the two variable case for $6 \leq N \leq 12$.

• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• 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
• /
• v.24 no.1
• /
• pp.31-44
• /
• 1995

In fitting a model, there always exists a discrepancy between the fitted model and the true functional relationship. In measuring this discrepancy, Box and Drapper (1959) used the criterion dividing the discrepancy into two parts which are the bias error part and the variance error one in single response case. In this paper, an optimum design which makes these two types of errors as small as possible is found by extending the Box and Drapper criterion to multiple response situation. Especially, a design is found to meat rotatability conditions when we fit a quadratic model to each response fearing cubic bias. Using the central composite design, an application of general results to a specific case is shown to help understanding the material.

• Journal of the Korean Statistical Society
• /
• v.17 no.2
• /
• pp.121-133
• /
• 1988

In this paper a class of mixture designs for estimating the slope of second order Scheffe polynomial response surfaces for mixture experiments with q components is presented. The variance of the estimated directional slope at a point is a function of the direction of the slope and the design. If the variance is averaged over all possible directions in the (q-1)-dimensional simplex, 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 centroid point. This property is called "slope-rotatability over al directions in the simplex", and the necessary and sufficient conditions for mixture design to have this property are given and proved. The class of designs with this property is compared with other mixture designs and discussed.discussed.