• Journal of Korean Society for Quality Management
• /
• v.22 no.1
• /
• pp.54-65
• /
• 1994

Multivariate exponentially weighted moving average (EWMA) control charts for monitoring the variance-covariance matrix are investigated. Two basic approaches, "combine-accumulate" approach and "accumulate-combine" approach, for using past sample information in the developement of multivariate EWMA control charts are considered. Multivariate EWMA control charts for monitoring the variance-covariance matrix are compared on the basis of their average run length (ARL) performances. The numerical results show that multivariate EWMA control charts based on the accumulate-combine approach are more efficient than corresponding multivariate EWMA control charts based on the combine-accumulate approach.

• Mass Spectrometry Letters
• /
• v.7 no.4
• /
• pp.91-95
• /
• 2016

In recent years, matrix-free laser desorption ionization (LDI) for mass spectrometry of thermally labile molecules has been an important research subject in the pursuit of new ionization methods to serve as alternatives to the conventional matrix-assisted laser desorption ionization (MALDI) method. While many recent studies have reported successful LDI of thermally labile molecules from various surfaces, mostly from surfaces with nanostructures, understanding of what drives the LDI process still requires further study. This article briefly reviews the thermal aspects involved in the LDI mechanism, which can be characterized as rapid surface heating. The thermal mechanism was supported by observed LDI and postsource decay (PSD) of peptide ions produced from flat surfaces with special thermal properties including amorphous Si (a-Si) and tungsten silicide ($WSi_x$). In addition, the concept of rapid surface heating further suggests a practical strategy for the preparation of LDI sample plates, which allows us to choose various surface materials including crystalline Si (c-Si) and Au tailorable to specific applications.

• Communications for Statistical Applications and Methods
• /
• v.13 no.1
• /
• pp.113-123
• /
• 2006

Normal mixture model(NMM) frequently used to cluster genes on microarray gene expression data. In this paper some of component means of NMM are modelled by a linear regression model so that its design matrix presents the pattern between sample classes in microarray matrix. This modelling for the component means by given design matrices certainly has an advantage that we can lead the clusters that are previously designed. However, it suffers from 'overfitting' problem because in practice genes often are highly dimensional. This problem also arises when the NMM restricted by the linear model for component-means is fitted. To cope with this problem, in this paper, the use of the factor analyzer NMM restricted by linear model is proposed to cluster genes. Also several design matrices which are useful for clustering genes are provided.

• Journal of the military operations research society of Korea
• /
• v.29 no.1
• /
• pp.76-87
• /
• 2003

This paper proposes that we need a simpler method for developing the Website strategy for public institutions. The research objectives are threefold: (1) A model that relates to the monitoring of Website strategy in the public sector is proposed. The model includes basic dimensions and a $2{\times}3$ matrix that is a simplified version of the Mohammed et dl.s Marketspace Matrix. (2) The model is tested empirically with a sample of 56 Websites selected from govemment agencies, military organizations and police stations in Korea. (3) The effect of dimension/factors on the innovation level is tested. A special attention is paid to online attracting that is important for public institutions which usually do not use offline advertising aggressively. Results from regression analyses show that main dimensions (Marketing Drivers and relationship Stages) and three factors (Basic Information, Support Information, and Participation) in the matrix are all significantly influential on the innovation level, but the Attracting is not. Colorful designs and attracting features of a homepage have not necessarily anything to do with innovation. This message can offer a good piece of advice for managers of Websites.

• Journal of the Korean Ceramic Society
• /
• v.34 no.10
• /
• pp.1092-1098
• /
• 1997

Electrical discharge machining (EDM) was attempted on a ceramic matrix composite containing non-conductive alumina as a matrix and conductive titania as a second phase, and was found successful. As the current or duty factor increased, the material removal rate (MRR) increased and the surface roughness also increased. The EDMed surface was covered with a number of craters of a circular shape having 100-200 microns of diameter. The melting and evaporation was suggested for the EDM mechanism. The bending strength decreased 44% after EDM, but the Weibull modulus increased more than twice. Combination of EDM and barre이 polishing resulted in the maintenance of the bending strength level. Temperature distribution near a spark in the sample was computer-simulated by use of finite element method, and was found to have similar shape to the one which the observed craters have.

• Structural Engineering and Mechanics
• /
• v.5 no.6
• /
• pp.785-801
• /
• 1997

Reinforced Concrete beams with tension and compression softening material constitutive laws are studied. Energy-based and non-local regularisation techniques are presented and applied to a R.C. element. The element characteristics (sectional tangent stiffness matrix, element tangent stiffness matrix restoring forces) are directly derived from their symbolic expressions through numerical integration. In this way the same spatial grid allows us to obtain a non-local strain estimate and also to sample the contributions to the element stiffness matrix. Three examples show the spurious behaviors due to the strain localization and the stabilization effects given by the regularisation techniques, both in the case of tension and compression softening. The possibility to overestimate the ultimate load level when the non-local strain measure is applied to a non softening material is shown.

• The Korean Journal of Applied Statistics
• /
• v.5 no.2
• /
• pp.283-292
• /
• 1992

Updating algorithms are studied for the basic statistics (mean, variance). For a linear model, a recursive formulae for least squares estimators of regression coefficients, residual sum of squares and variance-covariance matrix are also studied. Hotelling's $T^2$ statistics can be calculated recursively using the recursive formulae of mean vector and variance-covariance matrix without computing the sample variance-covariance matrix at each stage.

• Proceedings of the Korean Vacuum Society Conference
• /
• 1999.07a
• /
• pp.192-192
• /
• 1999

SIMS(Secondary Ion Mass Spectrometry)는 다른 표면 분석장비와 비교하여^g , pp m,^g , pp b 단위의 미량분석이 가능한 장비로서, 특히 depth Profiling을 위한 dynamic SIMS는 Mass Spectrometer의 종류에 따라 Quadrupole SIMS (Q-SIMS)와, Magnetic Sector SIMS (M-SIMS)로 분류된다. 한편, Q-SIMS와 달리 M-SIMS의 경우, Transmission을 높여 주기 위해 Sample Holder에 수 keV의 bias를 걸어 주는데, 이로 인하여 분석 원소에 대한 Sensitivity가 향상되어 지는 반면, RSF의 변화와 같은 분석상의 Artifact가 발생하게 된다. 일반적으로 Q-SIMS의 경우에는 RSF의 RSD(Relative Standard Deviation)가 1%이내에서 보고되고 있지만 M-SIMS에 있어서는 이러한 Deviation이 M-SIMS보다 크게 나타난다. 이 차이는 주로 Sample Holder와 Immersion Lens 사이에 형성되는 Magnetic Field의 왜곡과 Spectrometer의 문제로부터 발생한다. 본 논문에서는 Sample Holder의 종류 및 holder so window 위치에 따라 RSF의 차이를 측정하고 그 data를 RS/1 통계 Package를 이용하여 계량적으로 검증하였으며, 그 차이의 원인과 대책을 제시하고자 한다. 실험에 사용된 Sample은 Si(100) p-type Wafer에 Boron을 이온 주입하여 제작하였다. 이온 주입 장비는 Varian E-500HP이며, 5.0E13 ions/cm2의 dose양을 80keV의 Energy로 각각 7도와 22도의 Tilt와 Twist Angle로 이온 주입을 하였다. SIMS분석에 사용된 Sample Holder는 각각 3 Hole, 9 Hole Type HOlder이며, 분석은 Cameca IMS-6f를 사용하여 B에 대한 Matrix Peak으로 28Si++를 얻었다. 실험 결과 3 Hole Type Sample Holder의 경우 RSF의 RSD는 5.84%, 9Hle Type Sample Holder의 경우는 14.3%로 나타났으나 분석 Window의 위치에 따르는 Grouping을 실시한 결과, 3 Hole Type Sample Holder의 경우 1.2%, 9Hole Type Sample Holder의 경우 9.8%로 RSF의 변화가 감소하였다. 이러한 Deviation은 Sample Holder를 Mount시킬 때 세 개의 Screw를 이용하여 Immersion Lens와의 평형을 잡아주기 때문에 발생하며, 이 Munting을 정확히 해줌으로써 RSF의 변화를 줄일 수 있으나, 실제로 완벽한 Mounting이 불가능하기 때문에 RSF를 일정하게 하기 위해서는 Sample Holder so Window의 취치를 일정하게 설정한 후 분석을 실시해야 한다고 판단된다.

• Communications for Statistical Applications and Methods
• /
• v.3 no.2
• /
• pp.37-42
• /
• 1996

Maximum likelihood estimation of Cholesky decomposition is considered under normality assumption. It is shown that maximum liklihood estimation gives a Cholesky decomposition of the sample covariance matrix. The joint distribution of the maximum likelihood estimators is derived. The ussual algorithm for a Cholesky decomposition is shown to be equivalent to a maximumlikelihood estimation of a Cholesky root when the underlying distribution is a multivariate normal one.

• Transactions on Electrical and Electronic Materials
• /
• v.5 no.2
• /
• pp.81-85
• /
• 2004

High T$\_$c/ (Sm/Y)$\_$1.8/Ba$\_$2.4/Cu$\_$3.4/O$\_$7-$\delta$/ ［(Sm/Y)］ superconductor, a combination of Y and Sm(50% each), was systematically investigated by the zone melt growth process. A sample prepared by this method showed well-textured microstructure, and (Sm/Y)$_2$BaCuO$\_$5/［(Sm/Y)211］inclusions were uniformly dispersed in large (Sm/Y)Ba$_2$Cu$_3$O$\_$y/ ［(Sm/Y)123］matrix. The sample showed a sharp superconducting transition at 91 K. The magnetization measurements of the (Sm/Y)1.8 sample exhibited the enhanced flux pinning, compared with Y$\_$1.8/Ba$\_$2.4/Cu$\_$3.4/O$\_$7-$\delta$/(Y1.8) sample without Sm. Critical current densities of (Sm/Y) 1.8 sample was 3.5${\times}$10$^4$A/$\textrm{cm}^2$ at 1 T and 77 K.