• Journal of the Korea Society for Simulation
• /
• v.4 no.1
• /
• pp.25-36
• /
• 1995

We develop a variance reduction technique in one simulation experiment whose purpose is to estimate the parameters of a first-order linear model. This method utilizes the control variates obtained during the course of simulation run under Schruben and Margolin's method (S-M method). The performance of this method is shown to be similar in estimating the main effects, and to be superior to S-M method in estimating the overall mean response in the hospital simulation experiment. For the general case, we consider that a proposed method may yield a better result than S-M method if selected control variates are highly correlated with the response at each design point.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.31 no.1 s.256
• /
• pp.113-120
• /
• 2007

Design optimization of layered plates bonding process is conducted by considering uncertainties in a manufacturing process, in order to reduce the crack failure arising due to the residual stress at the surface of the adherent which is caused by different thermal expansion coefficients. Robust optimization is peformed to minimize the mean as well as its variance of the residual stress, while constraining the distortion as well as the instantaneous maximum stress under the allowable reliability limits. In this optimization, the dimension reduction (DR) method is employed to quantify the reliability such as mean and variance of the layered plate bonding. It is expected that the DR method benefits the optimization from the perspectives of efficiency, accuracy, and simplicity. The obtained robust optimal solution is verified by the Monte Carlo simulation.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2005.05a
• /
• pp.45-49
• /
• 2005

본 연구는 시뮬레이션 반응변수가 입력 인자의 선형 1차식으로 표현된 경우에 인자의 파라미터를 효과적으로 추정하기위해 사용될 수 있는 분산감소기법을 제안하였다. 이 기법은 하나의 실험설계에 공통난수와 대조난수를 동시에 사용하는 Schruben과 Margolin의 방법과 시뮬레이션하는 도중에 얻어지는 통제변수를 활용하는 기법을 결합하는 방법으로 시뮬레이션의 효율성을 개선하고자 하였다. 시뮬레이션 결과 제안된 기법은 주어진 모형의 평균 반응치를 추정한 데는 S-M 기법보다 효과적이었으며 인자의 다른 파라미터를 추정하는 데는 S-M 기법과 비슷한 성과를 보이고 있다. 만일 시뮬레이션 과정에서 반응변수와 상관성이 높은 통제변수들을 선택할 수 있는 경우에는 제안된 기법이 S-M 기법보다 보다 파라미터 추정에 효과적일 것으로 판단된다.

• Journal of the Korean Data Analysis Society
• /
• v.20 no.6
• /
• pp.2733-2746
• /
• 2018

The two inverse regression estimation methods, SIR and SAVE to estimate the central space are computationally easy and are widely used. However, SIR and SAVE may have poor performance in finite samples and need strong assumptions (linearity and/or constant covariance conditions) on predictors. The two non-parametric estimation methods, MAVE and dMAVE have much better performance for finite samples than SIR and SAVE. MAVE and dMAVE need no strong requirements on predictors or on the response variable. MAVE is focused on estimating the central mean subspace, but dMAVE is to estimate the central space. This paper explores and compares four methods to explain the dimension reduction. Each algorithm of these four methods is reviewed. Empirical study for simulated data shows that MAVE and dMAVE has relatively better performance than SIR and SAVE, regardless of not only different models but also different distributional assumptions of predictors. However, real data example with the binary response demonstrates that SAVE is better than other methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.11
• /
• pp.4355-4371
• /
• 2020

Phase retrieval, recovering a signal from phaseless measurements, is generally considered to be an NP-hard problem. This paper adopts an amplitude-based nonconvex optimization cost function to develop a new stochastic gradient algorithm, named stochastic reweighted phase retrieval (SRPR). SRPR is a stochastic gradient iteration algorithm, which runs in two stages: First, we use a truncated sample stochastic variance reduction algorithm to initialize the objective function. The second stage is the gradient refinement stage, which uses continuous updating of the amplitude-based stochastic weighted gradient algorithm to improve the initial estimate. Because of the stochastic method, each iteration of the two stages of SRPR involves only one equation. Therefore, SRPR is simple, scalable, and fast. Compared with the state-of-the-art phase retrieval algorithm, simulation results show that SRPR has a faster convergence speed and fewer magnitude-only measurements required to reconstruct the signal, under the real- or complex- cases.

• Journal of the Korean Society of Safety
• /
• v.12 no.4
• /
• pp.3-8
• /
• 1997

This paper shows a performance estimation of windowing a single tone with added Gaussian noise and uniform noise. Signal-to-noise ratio can be determined by the ratio of the output noisy signal variance to the input noisy signal variance of a window. Standard deviation of noise is reduced by windowing Signal-to-noise ratio of the noisy signal is reduced by the windowing operation. Thus, performance of window function can be determined by this filtering operation that improved the signal-to-noise ratio.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.31 no.6
• /
• pp.671-676
• /
• 2011

• Proceedings of the Korean Nuclear Society Conference
• /
• 2002.05a
• /
• pp.67-67
• /
• 2002

• Journal of Korean Society for Quality Management
• /
• v.22 no.4
• /
• pp.1-12
• /
• 1994

For quality improvement, Taguchi emphasizes the reduction of variation of the quality characteristic. Taguchi has used the signal to noise ratios for achieving minimum dispersion of the quality characteristic with its location adjusted to some desired target value. At each setting of design factors, the variance of the quality characteristic could be affected by the mean. In most cases, as the mean get larger, the variance tends to increase, The Taguchi's SN ratio corresponds to the case that the variance is proportional to the square of the mean. But the variance can increase faster or slower than the square of the mean. We propose to infer a linking relationship of the variance and mean through simple data analysis technique, and then use a reasonable SN ratio.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.42 no.2
• /
• pp.78-85
• /
• 2019

High variance observed in the measurement system can cause high process variation that can affect process capability badly. Therefore, measurement system analysis is closely related to process capability analysis. Generally, the evaluation for measurement system and process variance is performed separately in the industry. That is, the measurement system analysis is implemented before process monitoring, process capability and process performance analysis even though these analyses are closely related. This paper presents the effective concurrent evaluation procedure for measurement system analysis and process capability analysis using the table that contains Process Performance (Pp), Gage Repeatability & Reproducibility (%R&R) and Number of Distinct Categories (NDC). Furthermore, the long-term process capability index (Pp), which takes into account both gage variance and process variance, is used instead of the short-term process capability (Cp) considering only process variance. The long-term capability index can reflect well the relationship between the measurement system and process capability. The quality measurement and improvement guidelines by region scale are also described in detail. In conclusion, this research proposes the procedure that can execute the measurement system analysis and process capability analysis at the same time. The proposed procedure can contribute to reduction of the measurement staff's effort and to improvement of accurate evaluation.