• Industrial Engineering and Management Systems
• /
• v.14 no.1
• /
• pp.22-31
• /
• 2015

In this paper, a two-stage hybrid flow shop problem is considered. Specifically, there exist identical parallel machines at stage 1 and two dedicated machines at stage 2, and the objective of the problem is to minimize makespan. After being processed by any machine at stage 1, a job must be processed by a specific machine at stage 2 depending on the job type, and one type of jobs can have different processing times on each machine. First, we introduce the problem and establish complexity of several variations of the problem. For some special cases, we develop optimal polynomial time solution procedures. Then, we establish some simple lower bounds for the problem. In order to solve this NP-hard problem, three heuristics based on simple rules such as the Johnson's rule and the LPT (Longest Processing Time first) rule are developed. For each of the heuristics, we provide some theoretical analysis and find some worst case bound on relative error. Finally, we empirically evaluate the heuristics.

• Korean Management Science Review
• /
• v.15 no.1
• /
• pp.33-47
• /
• 1998

The facility location problem considered here is to determine facility location sites under future's uncertain demand. The objective of this paper is to propose a solution method and algorithm for a two-stage stochastic facility location problem. utilizing the Benders decomposition method. As a two-stage stochastic facility location problem is a large-scale and complex to solve, it is usually attempted to use a mean value problem rather than using a stochastic problem. Thus, the other objective is to study the relative error of objective function values between a stochastic problem and a mean value problem. The simulation result shows that the relative error of objective function values between two problems is relatively small, when a feasibility constraint is added to a facility location model.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.25 no.1
• /
• pp.1-8
• /
• 2011

In this paper, the method to calculate chromaticity coordinate from spectral power distribution of LED is presented. Also, inverse tri-stimulus algorithm to find mixed luminance of red, green, blue LED from targeted luminance and chromaticity coordinate is proposed. Besides, digital light color control system of LED lamp applied this algorithm has been developed. In experiments, each chromaticity coordinate of red, green, blue LED calculated from this algorithm has relative percentage error of few % to measured values. Digital code is drawn from inverse tri-stimulus algorithm, and measured values of luminance and chromaticity coordinate of LED lamp digitally controlled by this code also have relative percentage error within a few % to targeted luminance and chromaticity coordinate.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.27 no.3
• /
• pp.145-156
• /
• 2002

Business managers are often required to use LP problems to deal with uncertainty inherent in decision making due to rapid changes in today＇s business environments. Uncertain parameters can be easily formulated in the two-stage stochastic LP problems. However, since solution methods are complex and time-consuming, a common approach has been to use modified formulations to provide upper and lower bounds on the two-stage stochastic LP problem. One approach is to use an expected value problem, which provides upper and lower bounds. Another approach is to use “walt-and-see” problem to provide upper and lower bounds. The objective of this paper is to propose a modified approach of “wait-and-see” problem to provide an upper bound and to compare the relative error of optimal value with various upper and lower bounds. A computing experiment is implemented to show the relative error of optimal value with various upper and lower bounds and computing times.

• Journal of Industrial Technology
• /
• v.21 no.A
• /
• pp.41-50
• /
• 2001

A comparison is made of the temperature distribution and heat loss from a trapezoidal profile fin using two different 3-dimensional methods. These two methods are analytical and finite difference methods. In the finite difference method 78 nodes are used for a fourth of the fin. A trapezoidal profile fin being the height of the fin tip is half of that of the fin base is chosen arbitrarily as the model. One of the results shows that the relative error in the total convection heat loss obtained by using 78 nodes in the finite difference method as compared to the heat conduction through the fin root obtained by analytic method seems to be good (i.e., -3.5%

• Proceedings of the SAREK Conference
• /
• 2009.06a
• /
• pp.1177-1181
• /
• 2009

Heat loss from a convective rectangular profile annular fin with variable inside fluid heat transfer coefficient and fin height is calculated by using both the one dimensional analytic method and two dimensional variables separation method. Heat loss from the two dimensional method and the relative error of heat loss between the one dimensional method and two dimensional method are presented as a function of the fin length, ambient convection characteristic number and fin height. One of the results shows that the relative error of heat loss between one dimensional method and two dimensional method is within 0.7% in the range of given parameters in this study.

• Proceedings of the KIEE Conference
• /
• 2015.07a
• /
• pp.1431-1432
• /
• 2015

A system was developed to continuously monitor the respiratory signal under critical care, followed by performance evaluation for clinical application. Very much accurate pressure measurement was made in the endo-tracheal tube with mean relative error of 0.13%. In the comparative experiment simulating inspiration and expiration, the tidal volume measurement was also accurate within mean relative error of 3%, validating reliable measurement of respiratory signals.

• The Korean Journal of Applied Statistics
• /
• v.25 no.5
• /
• pp.733-741
• /
• 2012

The successful launch of Arirang satellites allow the acquisition of high resolution satellite imagery of Korean territory and enables the transition from the conventional cultivation area survey method to new image based methods adopted in advanced nations. In this study, we suggested reasonable sizes of the primary sampling unit and the secondary sampling unit for the satellite imagery based sampling design in 8 provinces preselected for this research. The PSU size was determined mainly in consideration of intracorrelation that shows the degree of homogeneity within each cluster and the efficiency of the image process. For the SSU size, we considered the relative standard error and the differences between the land cover maps produced by the Ministry of Environment and the satellite imagery processed by the National Statistical Office.

• Phonetics and Speech Sciences
• /
• v.6 no.2
• /
• pp.29-34
• /
• 2014

In the Vector Taylor Series (VTS)-based noisy speech recognition methods, Hidden Markov Models (HMM) are usually trained with clean speech. However, better performance is expected by training the HMM with noisy speech. In a previous study, we could find that Minimum Mean Square Error (MMSE) estimation of the training noisy speech in the log-spectrum domain produce improved recognition results, but since the proposed algorithm was done in the log-spectrum domain, it could not be used for the HMM adaptation. In this paper, we modify the previous algorithm to derive a novel mathematical relation between test and training noisy speech in the cepstrum domain and the mean and covariance of the Multi-condition TRaining (MTR) trained noisy speech HMM are adapted. In the noisy speech recognition experiments on the Aurora 2 database, the proposed method produced 10.6% of relative improvement in Word Error Rates (WERs) over the MTR method while the previous MMSE estimation of the training noisy speech produced 4.3% of relative improvement, which shows the superiority of the proposed method.

• Bulletin of the Korean Chemical Society
• /
• v.30 no.11
• /
• pp.2717-2722
• /
• 2009

The use of the classification and regression tree (CART) methodology was studied in a quantitative structure-activity relationship (QSAR) context on a data set consisting of the binding affinities of 39 imidazobenzodiazepines for the α1 benzodiazepine receptor. The 3-D structures of these compounds were optimized using HyperChem software with semiempirical AM1 optimization method. After optimization a set of 1481 zero-to three-dimentional descriptors was calculated for each molecule in the data set. The response (dependent variable) in the tree model consisted of the binding affinities of drugs. Three descriptors (two topological and one 3D-Morse descriptors) were applied in the final tree structure to describe the binding affinities. The mean relative error percent for the data set is 3.20%, compared with a previous model with mean relative error percent of 6.63%. To evaluate the predictive power of CART cross validation method was also performed.