• The Korean Journal of Applied Statistics
• /
• v.21 no.2
• /
• pp.275-290
• /
• 2008

The size of microarray gene expression data is very big and its observation process is also very complex. Thus missing values are frequently occurred. In this paper we propose the sequential partial least squares(SPLS) regression fitting method to estimate missing values for time course gene expression data that has correlations among observations over time points. The SPLS method is to combine the sequential technique with the partial least squares(PLS) regression fitting method. The usefulness of method proposed is evaluated through some simulation study for three yeast time course data.

• Journal of Mechanical Science and Technology
• /
• v.17 no.5
• /
• pp.698-707
• /
• 2003

A new quadratic response surface modeling method is presented. In this method, the incomplete small composite design (ISCD) is newly proposed to .educe the number of experimental runs than that of the SCD. Unlike the SCD, the proposed ISCD always gives a unique design assessed on the number of factors, although it may induce the rank-deficiency in the normal equation. Thus, the singular value decomposition (SVD) is employed to solve the normal equation. Then, the duality theory is used to newly develop the conservative least squares fitting (CONFIT) method. This can directly control the ever- or the under-estimation behavior of the approximate functions. Finally, the performance of CONFIT is numerically shown by comparing its'conservativeness with that of conventional fitting method. Also, optimizing one practical design problem numerically shows the effectiveness of the sequential approximate optimization (SAO) combined with the proposed ISCD and CONFIT.

• Communications for Statistical Applications and Methods
• /
• v.29 no.6
• /
• pp.721-733
• /
• 2022

In this paper we compare parameter estimation by Grassmann manifold optimization and sequential candidate set algorithm in a structured principal fitted component (PFC) model. The structured PFC model extends the form of the covariance matrix of a random error to relieve the limits that occur due to too simple form of the matrix. However, unlike other PFC models, structured PFC model does not have a closed form for parameter estimation in dimension reduction which signals the need of numerical computation. The numerical computation can be done through Grassmann manifold optimization and sequential candidate set algorithm. We conducted numerical studies to compare the two methods by computing the results of sequential dimension testing and trace correlation values where we can compare the performance in determining dimension and estimating the basis. We could conclude that Grassmann manifold optimization outperforms sequential candidate set algorithm in dimension determination, while sequential candidate set algorithm is better in basis estimation when conducting dimension reduction. We also applied the methods in real data which derived the same result.

• Proceedings of the KSME Conference
• /
• 2003.11a
• /
• pp.1334-1339
• /
• 2003

An efficient method is developed for the shape optimization of 2-D structures. The sequential linear programming is used for minimization problems. Selected set of master nodes are employed as design variables and assigned to move towards the normal direction. After adapting the nodes on the design boundary, the B-spline curves and mesh smoothing schemes are used to maintain the finite element in good quality. Finally, a numerical implementation of optimum design of an automobile torque converter piston subjected to pressure and centrifugal loads is presented. The results shows additional weight up to 13% may be saved after the shape optimization.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.33 no.9
• /
• pp.892-902
• /
• 2009

An efficient sequential optimization approach for metamodel was presented by Choi et al. This paper describes a new approach of the multilevel optimization method studied in Refs. [2] and [20,21]. The basic idea is concerned with multilevel iterative methods which combine a descent scheme with a hierarchy of auxiliary problems in lower dimensional subspaces. After fitting a metamodel based on an initial space filling design, this model is sequentially refined by the expected improvement criterion. The advantages of the method are that it does not require optimum sensitivities, nonlinear equality constraints are not needed, and the method is relatively easy to understand and use. As a check on effectiveness, the proposed method is applied to an engineering example.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10b
• /
• pp.1204-1209
• /
• 1990

This paper offers the theory and method for regression analysis of the regression model with operational parameter variables based on the fundamentals of mathematical statistics. Regression coefficients are usually constants related to the problem of regression analysis. This paper considers that regression coefficients are not constants but the functions of some operational parameter variables. This is a kind of method of two-step fitting regression model. The second part of this paper considers the experimental step numbers as recursive variables, the recursive identification with unknown operational parameter variables, which includes two recursive variables, is deduced. Then the optimization and the recursive identification are combined to obtain the sequential experiment optimum design with operational parameter variables. This paper also offers a fast recursive algorithm for a large number of sequential experiments.

• Journal of Korean Society for Quality Management
• /
• v.25 no.3
• /
• pp.96-107
• /
• 1997

The recent stream to reliability prediction is that it is totally inclusive in depth to consider even the operating and environmental condition at the level of finished goods as well as component itselves. In this study, firstly we present the reliability prediction methods by entire failure rate model which failure rate at the system level is added to the failure rate model at the component level. Secondly we build up the improved bases of reliability demonstration through a, pp.ication of Kaplan-Meier, Cumulative hazard, Johnson's methods as non-parametric and Maximum Likelihood Estimator under exponential & Weibull distribution as parametric. And also present the methods of curve fitting to piecewise failure rate under Weibull distribution, PRST (Probability Ratio Sequential Test), curve fitting to S-shaped reliability growth curve, computer programs of each methods. Lastly we show the practical for determination of optimal burn-in time as a method of reliability enhancement, and also verify the practical usefulness of the above study through the a, pp.ication of failure and test data during 1 year.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.1198-1203
• /
• 2008

An efficient sequential optimization approach for metamodel was presented by Choi et al [6]. This paper describes a new approach of the multilevel optimization method studied in Refs. [5] and [21-25]. The basic idea is concerned with multilevel iterative methods which combine a descent scheme with a hierarchy of auxiliary problems in lower dimensional subspaces. After fitting a metamodel based on an initial space filling design, this model is sequentially refined by the expected improvement criterion. The advantages of the method are that it does not require optimum sensitivities, nonlinear equality constraints are not needed, and the method is relatively easy to understand and use. As a check on effectiveness, the proposed method is applied to a classical cantilever beam.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.17 no.1
• /
• pp.31-38
• /
• 2004

Mechanical design process is generally accomplished by design, analysis, and test. Designers use programs fitting purpose, and obtain repeatedly a response of a simulation program, a sub-program for optimization. In this paper, shape optimization using approximate optimization technique is carried out with sequential design domain(SDD). In addition, algorithm executing Pro/Engineer and ANSYS automatically are adopted in the approximate optimization program by SDD. It is difficult for design problem to be approximated accurately for the whole range of design space. However, more or less accurate approximation is constructed if SDD is applied to that case. SDD starts with a certain range which is off-seted from midpoint of an initial design domain and then SDD of the next step is determined by a move limited. Convergence criterion is defined such that optimal point must be located within SDD during the two steps. Also, the PLBA(Pshenichny-Lim-Belegundu-Arora) algorithm is used to solve approximate optimization problems. This algorithm uses the second-order information and the active set strategy, in order to seek the direction of design variables.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1998.03a
• /
• pp.101-105
• /
• 1998

A large system predictor, which can perform prediction of sales trend in a huge number of distribution centers, is presented using neural predictive model. There are 20,000 number of distribution centers, and each distribution center need to forecast future demand in order to establish a reasonable inventory policy. Therefore, the number of forecasting models corresponds to the number of distribution centers, which is not possible to estimate that kind of huge number of accurate models in ERP (Enterprise Resource Planning)module. Multilayer neural net as universal approximation is employed for fitting the prediction model. In order to improve prediction accuracy, a sequential simulation procedure is performed to get appropriate network structure and also to improve forecasting accuracy. The proposed simulation procedure includes neural structure identification and virtual predictive model generation. The predictive model generation consists of generating virtual signals and estimating predictive model. The virtual predictive model plays a key role in tuning the real model by absorbing the real model errors. The complement approach, based on real and virtual model, could forecast the future demands of various distribution centers.