• The Korean Journal of Applied Statistics
• /
• v.7 no.2
• /
• pp.201-211
• /
• 1994

Ignoring interaction effects has been pointed out to be one of serious drawbacks in analysis of the parameter designs which are constructed by using orthogonal arrays. In this paper a detecting procedure of influential 2-factor inteactions with minimum expeimental runs is described, when each contrl factor has two levels. The presented method is based on the near orthogonal arrays which are very similar to orthogonal arrays in the statistical structure. And those arrarys are the same as trace-optimal balanced saturated two-level fractional factorial designs of resolution V.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.5
• /
• pp.29-36
• /
• 2009

Cost and Time for software test is gradually increasing as the software complexity increases. To cope with this problem, it is very important to reduce the number of test cases used in the software test. The interaction strength number is especially important in decision of the number of test cases for the unit test, where the interaction strength number means the number of arguments which affect the results of a function by the analysis of their combination used in source code of the function. This paper proposes the algorithm that computes the number of the interaction strength, where analyzes the patterns used in the source code of a function and increase its number when the pattern matches one of the specified patterns. The proposed algorithm is validated by some experiments finding coverage and the number of fault detection.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.382-384
• /
• 2005

소프트웨어 테스팅에서 테스트 스위트(suite)의 수를 줄이면서도 테스트 커버리지나 오류검출에 있어서 효과적인 방법을 찾기 위한 많은 연구가 시도되어 왔다. 안정성이 높은 설계가 되도록 설계 조건을 결정하기 위하여 제어인자들의 직교 배열을 사용하는 시험방법인 교호작용 테스팅 기법이 소프트웨어의 테스트에서도 효과가 매우 높다는 것이 실험으로 증명되고 있다. 소프트웨어는 상대적으로 몇 안 되는 조건들의 조합들로 오류가 발생할 가능성이 높다는 특징을 가지고 있다. 따라서, 파라미터 간 교호작용 강도 t를 갖는 t-way 테스팅을 통해 효과적으로 테스트 스위트를 줄이면서 많은 오류를 검출할 수 있다. 그러나 t값을 증가시키면 테스트 스위트의 수가 늘어난다는 단점이 발생한다. 또한, 어떠한 파라미터들이 서로 교호작용을 일으키는 것인지 알수 없는 상황을 가정한 단순한 교호작용 테스팅은 자칫 서로 관련 없는 파라미터들도 교호작용 테스팅에 참여하기 때문에 테스트의 낭비가 있을 수 있다는 단점이 있다. 이에, 본 논문에서는 소프트웨어의 입력과 출력간의 관계를 바탕으로 시스템 I-O관계도를 작성한 후, 이를 바탕으로 각 출력에 대한 교호작용을 일으킬 수 있는 파라미터를 중심으로 테스트 스위트를 생성하는 Selective Covering Array를 제안한다.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.43 no.4
• /
• pp.592-603
• /
• 2014

We developed various recipes of turmeric powder (Curcuma longa L.) added to milk bread and assessed the individual effects of seven ingredients [milk ($X_1$), turmeric powder ($X_2$), bread improver ($X_3$), fresh yeast ($X_4$), butter ($X_5$), sugar ($X_6$), and salt ($X_7$)] as well as the 2-way interaction effects of the ingredients on the sensory characteristics of breads using fractional factorial design method. The center and end points of each component were determined via literature review and multiple test baking. Seven trained sensory test panels evaluated the outside appearance (OA), inside appearance (IA), and flavor & texture (FT) of 38 breads using 46 items of sensory evaluation. Findings are as follows: for the OA, $X_1$ (P<0.05) and $X_4$ (P<0.0001) exhibited significant individual effects, whereas $X_1*X_7$, $X_2*X_5$, $X_3*X_6$, and $X_4*X_6$ indicated significant interaction effects (P<0.05). For the IA, $X_1$ (P<0.0001), $X_4$ (P<0.0001), $X_6$ (P<0.05), $X_2*X_4$ (P<0.05), and $X_3*X_6$ (P<0.01) showed individual and interaction effects, respectively. For the FT, $X_1$ and $X_2$ showed the most significant individual effect (P<0.0001), followed by $X_4$, $X_5$ and $X_6$ (P<0.05) in descending order. $X_4*X_7$ indicated the only significant interaction effect. We computed the magnitudes of the 2-way interaction effects of the ingredients with a distinct emphasis. Model equations predicting the levels of the ingredient effects on the breads were also provided via regression analyses. In summation, $X_4$ appeared to be the most significant component affecting the sensory characteristics based on its individual and 2-way interaction effects. Further, $X_6$, $X_1$, $X_2$, and $X_5$ indicated both individual and interaction effects. $X_3$ and X7 showed only interaction effects. The center point effect appeared to be unequivocal for whole sensory characteristics. Findings of the present study may provide insights into the selection of ingredients to derive an optimal model for turmeric powder-added bread using the response surface method hereafter.

• The Korean Journal of Applied Statistics
• /
• v.5 no.1
• /
• pp.19-28
• /
• 1992

In this paper, minimal balanced $2^t$ fractional factorial designs which permit the estimation of main effects and 2-factor interactions are developed by using a partially balanced array. Such designs are characterized by a minimum number of runs and some balancedness property of the variance-covariance matrix of the estimates. In addition to describing the designs, optimality criteria are discussed and the trace-optimal designs are presented. The proposed designs are especially useful in Taguchi method, where we need to investigate up to 2-factor interactions of the control factors.

• Journal of the korean Society of Automotive Engineers
• /
• v.11 no.6
• /
• pp.3-16
• /
• 1989

본 고에서는 품질향상 측면보다는 실험계획법 자체에 중점을 두고 이를 소개하고자 한다. 실험게획의 일차적 목적은 어느 의미에서는 모형화 즉 반응치에 미치는 각 인자의 주효과의 교호작용 관게를 규명하는 것이라고 볼 수 있으며 방법론에서 전조합실시법과 부분실시법으로 구분할 수 있는데 부분실시법의 목적은 어떻게 하면 경제적으로 시험횟수를 최소화하되 전조합실시법의 효과를 낼 수 있는가 하는 것이다. Taguchi의 실험계획은 직교배열을 이용한 부분실시법에 근거를 두고 있다.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.5
• /
• pp.787-796
• /
• 2009

A random coefficient model is applied when times of the repeated measurements are not fixed in experiments with respect to the subjects. The procedures of the inference of a random coefficient model are same as those of a mixed model. Diet experimental data was used for applying the random coefficient model. Various random coefficient models are investigated for the experimental data, and are compared each other. Finally, optimal random coefficient model would be selected. It resulted from the analysis that for the fixed effect factor, the baseline, treatment, height, and time effect were very significant. The treatment effect of the diet foods and exercises were more effective in losing weight than the effect of the diet foods only. The fixed cubic time effect was very significant. The variance components corresponding to the subject effect, linear time effect, quadratic time effect, and cubic time effect of the random coefficients are all positive. When quartic time effect was added as random coefficients the model did not converge. Thus random coefficients up to the cubic terms was considered as the optimal model.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.21 no.2
• /
• pp.154-158
• /
• 2013

Expansion joint has been utilized in many areas including automotive bellows for exhaust system. Usage of expansion joint has been increased due to its inherent flexibility and excellent anti-vibration property. Simple shape of expansion joint is modeled to understand the behavior of joint system. 27 design cases using 3 design factors with 3 levels are constructed by design of experiment. Each case is simulated to find the most influential design factors. Response for this study, maximum stress in the expansion joint, has been used to determine main design factors of joint. Among the 3 design factors, factor B has affected greatly a response in the formation of optimum shape of joint. Also, interaction factor, $A{\times}B$, has also showed its influence to the response of joint. This study showed that design of experiment combined with finite element analysis could be used in the design decision process effectively in the design of expansion joint.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.37 no.3
• /
• pp.287-294
• /
• 2013

The roll-to-roll (R2R) printed electronics system is one of the most promising technologies for the printed electronics industry because of several advantages in terms of productivity and cost. In the R2R printed electronics system, the characteristics of the printed patterns are an important issue that determines the functional quality of the printed matter. This study analyzed how several main factors may affect the characteristics of printed patterns, especially the thickness and surface roughness. The statistical model for estimation of the printed pattern was developed as a function of the main factors using the design of experiment (DOE) methodology. Based on the statistical analysis results, the R2R printed electronics system can be designed to control the characteristics of printed patterns.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1263-1270
• /
• 2010

Sixty patients were divided into three groups. Each group of twenty persons had fed on different diet foods over 5 weeks. Cholesterol had been measured repeatedly five times at an interval of a week during 5 weeks. It resulted from mixed model analysis of repeated measurements data that homogeneous toeplitz covariance matrix pattern was selected as the optimal covariance pattern. The correlations between measurements of different times for the covariance matrix are somewhat highly correlated as 0.64-0.78. Based upon the homogeneous toeplitz covariance pattern model, the time effect was found to be highly significant, but the treatment effect and treatment-time interaction effect were found to be insignificant.