• 산업공학
• /
• 제23권3호
• /
• pp.229-238
• /
• 2010

Virtual metrology (VM), a critical component of semiconductor manufacturing, is an efficient way of assessing the quality of wafers not actually measured. This is done based on a model between equipment sensor data (obtained for all wafers) and the quality characteristics of wafers actually measured. This paper considers principal component regression (PCR), partial least squares regression (PLSR), kernel PCR (KPCR), and kernel PLSR (KPLSR) as VM models. For each regression model, two cases are considered. One utilizes all explanatory variables in developing a model, and the other selects significant variables using the genetic algorithm (GA). The prediction performances of 8 regression models are compared for the short- and long-term etch process data. It is found among others that the GA-KPLSR model performs best for both types of data. Especially, its prediction ability is within the requirement for the short-term data implying that it can be used to implement VM for real etch processes.

• 품질경영학회지
• /
• 제24권1호
• /
• pp.32-43
• /
• 1996

In this paper we develop nonparametric methods for regression analysis when the response variable is subject to censoring that arises naturally in quality engineering. This development is based on a general missing information principle that enables us to apply, via an iterative scheme, nonparametric regression techniques for complete data to iteratively reconstructed data from a given sample with censored observations. In particular, additive regression models are extended to right-censored data. This nonparametric regression method is applied to a simulated data set and the estimated smooth functions provide insights into the relationship between failure time and explanatory variables in the data.

• International journal of advanced smart convergence
• /
• 제13권1호
• /
• pp.206-211
• /
• 2024

This study used a total of 205,565 datasets of 'voltage', 'current', '℃', and 'time(s)' to systematically analyze the properties and performance of solid electrolytes. As a method for characterizing solid electrolytes, a linear regression model, one of the machine learning models, is used to visualize the relationship between 'voltage' and 'current' and calculate the regression coefficient, mean squared error (MSE), and coefficient of determination (R^2). The regression coefficient between 'Voltage' and 'Current' in the results of the linear regression model is about 1.89, indicating that 'Voltage' has a positive effect on 'Current', and it is expected that the current will increase by about 1.89 times as the voltage increases. MSE found that the mean squared error between the model's predicted and actual values was about 0.3, with smaller values closer to the model's predictions to the actual values. The coefficient of determination (R^2) is about 0.25, which can be interpreted as explaining 25% of the data.

• Journal of the Korean Data and Information Science Society
• /
• 제26권3호
• /
• pp.561-568
• /
• 2015

이 연구에서는 공간회귀모형 중 공간시차모형과 공간오차모형을 이용하여 대구 경북 지역 단위면적당 아파트 매매가격을 예측하였다. k-최근접이웃 (k-nearest neighbours)을 이용하여 공간가중행렬을 구축하였으며, 이를 이용해 2012년 3월의 단위면적당 아파트 매매가격에 대한 모형을 적합시켰다. 적합시킨 공간시차모형, 공간오차모형을 이용하여 2013년 3월의 단위면적당 아파트 매매가격을 예측하였으며 RMSE (root mean squared error), RRMSE (root relative mean squared error), MAE (mean absolute error)를 통해 두 모형의 성능을 비교하였다.

• 응용통계연구
• /
• 제18권3호
• /
• pp.701-712
• /
• 2005

로지스틱 회귀모형에서 suppression의 논의는 선형회귀의 논의보다 많지 않은데 그 이유 중의 하나는 회귀제곱합 또는 결정계수의 정의가 유일하지 않고 다양하기 때문이다. 여러 종류의 결정계수들 중에서 선호되는 두 종류의 결정계수와 Liao와 McGee(2003)가 제안한 두 종류의 수정 결정계수의 정의로부터 회귀제곱합을 유도하여 로지스틱 회귀모형에서의 suppression을 설명하고자 한다. 모의실험을 통하여 자료를 생성하여 어떤 경우에 suppression이 발생하는지를 살펴보고 그 결과를 선형회귀모형에서의 suppression 결과와 비교한다.

• Computers and Concrete
• /
• 제21권5호
• /
• pp.559-567
• /
• 2018

In this study alkali activation of Electric Arc Furnace Slag (EAFS) is studied with a comprehensive test program. Three different silicate moduli (1-1,5-2), three different sodium concentrations (4%-6%-8%) for each silicate module, two different curing conditions (45%-98% relative humidity) for each sodium concentration, two different curing temperatures ($400^{\circ}C-800^{\circ}C$) for each relative humidity condition and two different curing time (6h-12h) for each curing temperature variables are selected and their effects on compressive strength was evaluated then regression equations using multiple linear regressions methods are fitted. And then to select the best regression models confirm with using the variables, the regression models compared between itself. An Artificial Neural Network (ANN) models that use silicate moduli, sodium concentration, relative humidity, curing temperature and curing time variables, are formed. After the investigation of these ANN models' results, ANN and multiple linear regressions based models are compared with each other. After that, an explicit formula is developed with values of the ANN model. As a result of this study, the fluctuations of data set of the compressive strength were very well reflected using both of the methods, multiple linear regression with quadratic terms and ANN.

• 응용통계연구
• /
• 제22권3호
• /
• pp.515-530
• /
• 2009

Evaporation from surface water bodies is influenced by a number of meteorological parameters. The rate of evaporation is primarily controlled by incoming solar radiation, air and water temperature and wind speed and relative humidity. In the present study, influence of weekly meteorological variables such as air temperature, relative humidity, bright sunshine hours, wind speed, wind velocity, rainfall on rate of evaporation has been examined using 35 years(1971-2005) of meteorological data. Statistical analysis was carried out employing linear regression models. The developed regression models were tested for goodness of fit, multicollinearity along with normality test and constant variance test. These regression models were subsequently validated using the observed and predicted parameter estimates with the meteorological data of the year 2005. Further these models were checked with time order sequence of residual plots to identify the trend of the scatter plot and then new standardized regression models were developed using standardized equations. The highest significant positive correlation was observed between pan evaporation and maximum air temperature. Mean air temperature and wind velocity have highly significant influence on pan evaporation whereas minimum air temperature, relative humidity and wind direction have no such significant influence.

• Communications for Statistical Applications and Methods
• /
• 제15권2호
• /
• pp.229-234
• /
• 2008

A Cook-type distance is considered for investigating the influence of observations in constrained regression models. Its exact sampling distribution is derived, which is used for judging whether each observation is influential or not. A numerical example is provided for illustration.

• ETRI Journal
• /
• 제21권3호
• /
• pp.6-15
• /
• 1999

Many studies in the software reliability have attempted to develop a model for predicting the faults of a software module because the application of good prediction models provides the optimal resource allocation during the development period. In this paper, we consider the change request data collected from the field test of the software module that incorporate a functional relation between the faults and some software metrics. To this end, we discuss the general aspect if regression method, the problem of multicollinearity and the measures of model evaluation. We consider four possible regression models including two stepwise regression models and two nonlinear models. Four developed models are evaluated with respect to the predictive quality.

• 설비공학논문집
• /
• 제12권7호
• /
• pp.688-696
• /
• 2000

A scheme for fault detection on the subsystem level is presented. The method uses analytical redundancy and consists in generating residuals by comparing each measurement with an estimate computed from the reference models. In this study regression neural network models are used as reference models. The regression neural network is memory-based feed forward network that provides estimates of continuous variables. The simulation result demonstrated that the proposed method can effectively detect faults in an air handling unit(AHU). The results show that the regression models are accurate and reliable estimators of the highly nonlinear and complex AHU.