• Journal of Korean Society for Quality Management
• /
• v.29 no.1
• /
• pp.24-40
• /
• 2001

Response surface designs can be classified, according to the shape of the experimental region, into spherical designs and cuboidal designs. Among the central composite design(CCD)s and the Box-Behnken design(BBD)s that are popular in practice, when the number of factors is k, spherical designs are tile CCDs with the axial value being $\sqrt{\textit{k}}$ and the BBDs, and cuboidal designs are the CCDs with the axial value being 1. With the CCDs having $\sqrt{\textit{k}}$ as the axial value, the radius of the experimental region varies with number of factors, but these designs are the 5-level designs. With the BBDs that are 3-level designs, the radius of the experimental region does not vary with the number of factors. In this article, we propose tile 3-level spherical designs which are constructed so that tile radius of the experimental region varies with the number of factors.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.893-904
• /
• 2011

The lecture of the experimental designs consists of two main part-experimental designs and model analysis. Mostly, the progress of the visualization has been made on a model analysis. As the visualization of experimental designs, we can consider the visualization of Latin squares, supersaturated designs, and balanced incomplete block designs. We can propose the design plots as well as use the scatterplots and the scatterplot matrices for the visualization of experimental designs. Through the visualization of experimental designs, we can use the synergy effect in teaching the lecture of the experimental designs.

• IE interfaces
• /
• v.23 no.2
• /
• pp.118-125
• /
• 2010

The determination of a regression model is important in using statistical designs of experiments. Generally, the exact regression model is not known, and experimenters suppose that a certain model form will be fit. Then an experimental design suitable for that predetermined model form is selected and the experiment is conducted. However, the initially chosen regression model may not be correct, and this can result in undesirable statistical properties. We develop model-robust experimental designs that have stable prediction variance for a family of candidate regression models over a cuboidal region by using genetic algorithms and the desirability function method. We then compare the stability of prediction variance of model-robust experimental designs with those of the 3-level face centered cube. These model-robust experimental designs have moderately high G-efficiencies for all candidate models that the experimenter may potentially wish to fit, and outperform the cuboidal design for the second-order model. The G-efficiencies are provided for the model-robust experimental designs and the face centered cube.

• Journal of Korean Society for Quality Management
• /
• v.45 no.4
• /
• 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.

• Communications for Statistical Applications and Methods
• /
• v.16 no.1
• /
• pp.195-208
• /
• 2009

A conventional single design optimality criterion has been used to select an efficient experimental design. But, since an experimental design is constructed with respect to an optimality criterion pre specified by investigators, an experimental design obtained from one optimality criterion which is superior to other designs may perform poorly when the design is evaluated by another optimality criterion. In other words, none of these is entirely satisfactory and even there is no guarantee that a design which is constructed from using a certain design optimality criterion is also optimal to the other design optimality criteria. Thus, it is necessary to develop certain special types of experimental designs that satisfy multiple design optimality criteria simultaneously because these multi-optimal designs (MODs) reflect the needs of the experimenters more adequately. In this article, we present a heuristic approach to construct second-order response surface designs which are more flexible and potentially very useful than the designs generated from a single design optimality criterion in many real experimental situations when several competing design optimality criteria are of interest. In this paper, over cuboidal design region for $3\;{\leq}\;k\;{\leq}\;5$ variables, we construct multi-optimal designs (MODs) that might moderately satisfy two famous alphabetic design optimality criteria, G- and IV-optimality criteria using a GA which considers a certain amount of randomness. The minimum, average and maximum scaled prediction variances for the generated response surface designs are provided. Based on the average and maximum scaled prediction variances for k = 3, 4 and 5 design variables, the MODs from a genetic algorithm (GA) have better statistical property than does the theoretically optimal designs and the MODs are more flexible and useful than single-criterion optimal designs.

• Journal of the Korea Safety Management & Science
• /
• v.7 no.5
• /
• pp.69-84
• /
• 2005

This paper proposes statistically designed experiments which provide a proactive means to implement safety and environmental applications. Minimal experimental designs such as fractional factorial design, Plackett-Burman design, Box-Behnken design are economical and can be achieved tremendous savings with relatively few experiments. These experimental designs and analysis methods are illustrated with cases.

• Journal of the Korea Furniture Society
• /
• v.23 no.1
• /
• pp.1-9
• /
• 2012

Sustainable furniture designs are expressed through reutilization, reuse or redesign of the whole of part of natural materials and existing products. These designs contain natural emotions and show those propensities such as uncertainty, integrity, essentiality and experimental natures. Designs that use natural materials such as wood, bamboo, cork, plants, stones and water have sufficient beauty in themselves and thus these materials are not specially processed but their essence is pursued in those designs. Of those materials, the natural material water is used to pursue the essence of aqua while presenting a new furniture design through an experimental method. The circularity of aqua shares its meaning with the circularity of sustainable designs. In addition, the liquidity of aqua will be grafted onto IT technology to express variable shapes in order to present a new direction of furniture designs.

• The Korean Journal of Applied Statistics
• /
• v.10 no.1
• /
• pp.163-175
• /
• 1997

Randomization of the run order within a block is a technique commonly employed by the experimenters of block designs to avoid biases in the estimates of the effects of interest. In practice, however the experimental responses are sometimes affected by the spatial or temporal position of the experimental units within a block. In such cases, it is preferable to use a systematic ordering of the treatments. It is often possible to find an ordering which will allow the estimation of treatment effects independently of any trend is known as a trend-free block designs. In many idustrial and agricultural experiments, treatments are applied to experimental units sequentially in time or space. This paper begins with a review of concepts and properties of trend-free designs. We, then devise algorithms to generate linear trend-free designs. We extend and modify the existing algorithm which is given by Bradley and Odeh(1988). Also, the algorithm which generate all possible linear trend-free designs in provided.

• Korean Journal of Biological Psychiatry
• /
• v.19 no.4
• /
• pp.159-163
• /
• 2012

An appropriate research design for hypotheses and purposes leads to a good quality of research results. In this review article, we summarized the types of research methods and described the characteristics of clinical trials. Research designs are categorized into observational studies and experimental ones, depending on data collecting methods. In an observational study, there are cross-sectional, cohort and case-control studies. Parallel groups design and crossover trial studies are representative designs in a randomized controlled trial study, a kind of experimental study. Clinical researchers should understand the characteristics of clinical research designs including advantages and disadvantages and choose the suitable design according to their study purposes and the nature of collected data or subjects.

• The Korean Journal of Applied Statistics
• /
• v.13 no.2
• /
• pp.477-488
• /
• 2000

A distance measure between two Latin-hypercube designs is defined and its expected value is computed. It was computed by using mathematical statistics, numerical analysis (multidimensional numerical integration), Monte-carlo method, and the theory of asymptotic normal distribution. For the comparison of two Latin-hypercube designs with same structure but different randomness, the difference of expected values of response function and information mass of experimental designs are considered. These methods may be useful in comparison between two general experimental designs.