• Proceedings of the Korean Statistical Society Conference
• /
• 2001.11a
• /
• pp.109-114
• /
• 2001

m=2 또는 n=2이고, ${\lambda}_1<{\lambda}_2$인 그룹분류가능계획을 매개디자인으로 사용한 완전이면교배가 A-최적, D-최적임을 보였다. 또한, ${\lambda}_2={\lambda}_1+1$이면 일반화된 최적계획이 됨을 보였다.

• The Korean Journal of Applied Statistics
• /
• v.4 no.2
• /
• pp.171-178
• /
• 1991

For the polynomial spline regression with fixed knots, some properties of the D-optimal design are discussed. Also the D-optimal design for some cases are found analytically by using a normalized B-spline basis for $S(P_m : k : \Delta)$. Based on the Kiefer-Wolfowitz equivalence theorem, the D-optimal design for some cases are found by numerical methods.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.5
• /
• pp.535-541
• /
• 2015

In this study, an approximate multi-objective optimization of a wall-mounted monitor bracket arm was performed. The rotation angle of the bracket arm was determined considering the inplane degree of freedom. We then formulated an optimization problem on maximum stress and deflection. Analyses of mean and design parameters were conducted for sensitivity regarding performance with orthogonal array and response surface method (RSM). RSM models of objective and constraint functions were generated using central composite (CCD) and D-optimal design. The accuracy of approximate models was evaluated through $R^2$ value. The obtained optimal solutions by non-dominant sorting genetic algorithm (NSGA-II) were validated through the finite element analysis and we compared the obtained optimal solution by CCD and D-optimal design.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.47 no.5
• /
• pp.379-387
• /
• 2019

The complexity of path planning for visiting multiple mission points is even larger than that of single pair path planning. Deciding a path for visiting n mission points requires conducting $n^2+n$ times of single pair path planning. We propose Multiple Mission $D^*$ Lite($MMD^*L$) which is an optimal path planning algorithm for visiting multiple mission points in dynamic environments. $MMD^*L$ reduces the complexity by reusing the computational data of preceding single pair path planning. Simulation results show that the complexity reduction is significant while its path optimality is not compromised.

• The Korean Journal of Applied Statistics
• /
• v.13 no.2
• /
• pp.393-405
• /
• 2000

Box and Draper( 19(5) listed some properties of a design that should be considered in design selection. But it is impossible that one design criterion from optimal experimental design theory reflects many potential objectives of an experiment, because the theory was originally based on the underlying model and its strict assumption about the error structure. Therefore, when it is neces::;ary to implement multi-objective experimental design. it is common practice to balance out the several optimal design criteria so that each design criterion involved benefits in terms of its relative "high" efficiency. In this study, we proposed several composite design criteria taking the case of heteroscedastic model. WVhen the heteroscedasticity is present in the model. the well known equivalence theorem between 1)- and C-optimality no longer exists and furthermore their design characteristics are sometimes drastically different. We introduced three different design criteria for this purpose: constrained design, combined design, and minimax design criteria. While the first two methods do reflect the prior belief of experimenter, the last one does not take it into account. which is sometimes desirable. Also we extended this method to the case when there are uncertainties concerning the error structure in the model. A simple algorithm and concluslOn follow.On follow.

• The Korean Journal of Applied Statistics
• /
• v.10 no.1
• /
• pp.69-84
• /
• 1997

We consider a linear calibration problem, $y_i = $$\alpha + \beta (x_i - x_0) + \epsilon_i$, $i=1, 2, {\cdot}{\cdot},n$ $y_f = \alpha + \beta (x_f - x_0) + \epsilon, $ where we observe $(x_i, y_i)$'s for the controlled calibration experiments and later we make inference about $x_f$ from a new observation $y_f$. The objective of the calibration design problem is to find the optimal design $x = (x_i, \cdots, x_n$ that gives the best estimates for $x_f$. We compare Kim(1989)'s Bayesian design which minimizes the expected value of the posterior variance of $x_f$ and some optimal designs from literature. Kim suggested the Bayesian optimal design based on the analysis of the characteristics of the expected loss function and numerical must be equal to the prior mean and that the sum of squares be as large as possible. The designs to be compared are (1) Buonaccorsi(1986)'s AV optimal design that minimizes the average asymptotic variance of the classical estimators, (2) D-optimal and A-optimal design for the linear regression model that optimize some functions of $M(x) = \sum x_i x_i'$, and (3) Hunter & Lamboy (1981)'s reference design from their paper. In order to compare the designs which are optimal in some sense, we consider two criteria. First, we compare them by the expected posterior variance criterion and secondly, we perform the Monte Carlo simulation to obtain the HPD intervals and compare the lengths of them. If the prior mean of $x_f$ is at the center of the finite design interval, then the Bayesian, AV optimal, D-optimal and A-optimal designs are indentical and they are equally weighted end-point design. However if the prior mean is not at the center, then they are not expected to be identical.In this case, we demonstrate that the almost Bayesian-optimal design was slightly better than the approximate AV optimal design. We also investigate the effects of the prior variance of the parameters and solution for the case when the number of experiments is odd.

• The Korean Journal of Applied Statistics
• /
• v.4 no.2
• /
• pp.195-207
• /
• 1991

We consider the development of simple regression-like diagnostics for assessing the sensitivity of an optimal design to deviations from the assumptions of constant error variance. This contains a review of Cook's local influence approach and an application of local influence aproach to D-optimal experimental design. The method is applied in a number of simple examples in Section 3. Conclusions and directions for further research follow in Section 4.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.7
• /
• pp.673-679
• /
• 2015

In this study, the nondominated sorting genetic algorithm (NSGA-II) is used to obtain the optimized proportional-integral-derivative (PID) gain value that can quickly recover the motion of a quadcopter after a disturbance. Prior to PID control, the four-rotor quadcopter interval was defined using computational fluid dynamics (CFD). Through the definition of this model, the PID control algorithm was generated. To construct a response surface model, D-optimal programming was used for the generation of experimental points. For this purpose, a gain value that satisfies both the roll and altitude PID gain values is obtained. Using the NSGA-II, the gain value of shorten time of the quadcopter motion control can be optimized.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.4
• /
• pp.765-772
• /
• 2011

In this paper, a simulation-based method is presented to dispose direction finders in three dimensional space for locating targets using the directional data. A direction finder(DF) is a military weapon that is used to find locations of targets that emit radio frequencies by operating two or more DFs simultaneously. If one or more DFs are operated in the air, the accuracy of location estimation can be enhanced by disposing them in a better configuration. By extending the line method, which is a well-known algorithm for 2-D location estimation, into 3-D space, the problem of 3-D location estimation is defined as an nonlinear programming form and solved analytically. Then the optimal disposition of DFs is considered with the presented method in which methods of simulation and search technique are combined. With the suggested algorithm for 3-D disposition of DFs, regions in which targets exist can be effectively covered so that the operation effect of DF be increased.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.4 s.175
• /
• pp.894-900
• /
• 2000

Optimization of chassis frame is performed according to the minimization of eleven responses representing one total frame weight, three natural frequencies and seven strength limits of chassis frame that are analyzed by using each response surface model from D-optimal design of experiments. After each response surface model is constructed form D-optimal design and random orthogonal array, the main effect and sensitivity analyses are successfully carried out by using this approximated regression model and the optimal solutions are obtained by using a nonlinear programming method. The response surface models and the optimization algorithms are used together to obtain the optimal design of chassis frame. The eleven-polynomial response surface models of the thirteen frame members (design factors) are constructed by using D-optimal Design and the multi-disciplinary design optimization is also performed by applying dual response analysis.