• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.04a
• /
• pp.120-120
• /
• 1996

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.32 no.1
• /
• pp.117-129
• /
• 2009

The bullwhip effect is known as the significant factor which causes unnecessary inventory, lost sales or cost increase in supply chains. Therefore, the causes of the bullwhip effect must be examined and removed. In this paper, we develop two analytical tools for the bullwhip effect control in supply chains. First, we develop the quantitative models for computing the bullwhip effect in a three-stage supply chain consisted of a single retailer, a single distributor and a single manufacturer when the fixed-interval replenishment policy is applied at each stage. The quantitative models are developed under the different conditions for the demand forecasting and share of customer demand information. They are validated through the computational experiments. Second, we develop a simulation-based decision support system for the bullwhip effect control in a more diverse dynamic supply chain environment. The system includes a what-if analysis function to examine the effects of varying input parameters such as operating policies and costs on the bullwhip effect.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.12
• /
• pp.3274-3292
• /
• 2013

Unsupervised methods for image segmentation are recently drawing attention because most images do not have labels or tags. A topic model is such an unsupervised probabilistic method that captures latent aspects of data, where each latent aspect, or a topic, is associated with one homogeneous region. The results of topic models, however, usually have noises, which decreases the overall segmentation performance. In this paper, to improve the performance of image segmentation using topic models, we propose two topic masks applicable to topic assignments of homogeneous regions obtained from topic models. The topic masks capture the noises among the assigned topic assignments or topic labels, and remove the noises by replacements, just like image masks for pixels. However, as the nature of topic assignments is different from image pixels, the topic masks have properties that are different from the existing image masks for pixels. There are two contributions of this paper. First, the topic masks can be used to reduce the noises of topic assignments obtained from topic models for image segmentation tasks. Second, we test the effectiveness of the topic masks by applying them to segmented images obtained from the Latent Dirichlet Allocation model and the Spatial Latent Dirichlet Allocation model upon the MSRC image dataset. The empirical results show that one of the masks successfully reduces the topic noises.

• Nuclear Engineering and Technology
• /
• v.32 no.6
• /
• pp.618-635
• /
• 2000

A total of 17 experimental data for the onset of slugging, which is assumed to be the precursor of the condensation-induced waterhammer (CIWH), have been obtained for various How rates of water Incorporating the most recent correlations of interfacial heat transfer and friction factor developed for a circular geometry and using an improved criterion of transition from stratified to a slug flow, two existing analytical models to predict lower and upper bounds for CIWH have been upgraded. Applicability of the present as well as existing CIWH models has been tested by comparison with two sets of CIWH data. The result of this comparison shows that the applicability of the present as well as existing models is reasonably good. Based on the present models for CIWH, a computer code entitled as“KAIST-CIWH”has been developed and sample guide charts to find CIWH free regions for a given combination of major flow parameters in a long horizontal pipe have been presented along with the results of parametric studies of major parameters (D, P, $T_{f,in}$, and L/D) on the critical inlet water flow rate（$W_{f,in}_crit$ for both lower and upper bounds. In addition, two simple formulas for lower and upper bounds that can be used in an emergency for quick results have been presented.

• Journal of the Korean Electrochemical Society
• /
• v.8 no.2
• /
• pp.106-114
• /
• 2005

This article covers the theoretical ac-impedance models for the analysis of oxygen reduction on the porous cathode electrode f3r solid oxide fuel cell (SOFC). Firstly, ac-impedance models were explained on the basis of the mechanism of oxygen reduction, which were classified into the rate-determining steps; (i) adsorption of oxygen atom on the electrode surface, (ii) diffusion of adsorbed oxygen atom along the electrode surface towards the three-phase (electrode/electrolyte/gas) boundaries, (iii) surface diffusion of adsorbed oxygen atom m ixed with the adsorption reaction of oxygen atom on the electrode surface and (iv) diffusion of oxygen vacancy through the electrode coupled with the charge transfer reaction at the electrode/gas interface. In each section for ac-impedance model, the representative impedance plots and the interpretation of important parameters attributed to the oxygen reduction reaction were explained. Finally, we discussed in detail the applications of the proposed theoretical ac-impedance models to the real electrode of SOFC system.

• Nuclear Engineering and Technology
• /
• v.31 no.5
• /
• pp.465-475
• /
• 1999

The objective of the present work is to assess the analysis capability of two wall film condensation models, the default and the alternative models, of RELAP5/MOD3.2 on condensation experiments in the presence of noncondensable gas in a vertical tube of PCCS of CP-1300. In the calculation of a base case the default model of RELAP5/MOD3.2 under-predicts the heat transfer coefficients, and Its alternative model over-predicts them throughout the condensing tube, Also, both models over-predict the void fractions. The nodalization study shows that the variation of the node number does not change both modeling results of RELAP5/MOD3.2 Sensitivity study for varying input parameters shows that the inlet steam-air mixture flow rate, the inlet air mass fraction, and the inlet saturated steam temperature give significant changes of their heat transfer coefficients Run statistics show that the grind time of the default model is always higher than that of the alternative model by about 23%.

• Journal of the Korea Safety Management & Science
• /
• v.6 no.1
• /
• pp.257-268
• /
• 2004

According to the World Competitiveness Yearbook(WCY), competitiveness of nations looks how nations create and maintain an environment which sustains the competitiveness of its enterprises. To develop economics and join the ranks of the advanced countries in the global market, Korea needs to consider the national competitiveness. Consequently, in this study, first, we try to know a concept of national competitiveness. Second, we also investigate models of national competitiveness. Finally, we suggest an improvement strategy, which is based on models, as compared with many countries of Europe.

• Bulletin of the Korean Mathematical Society
• /
• v.45 no.4
• /
• pp.709-715
• /
• 2008

We construct certain ring class fields over an imaginary quadratic field by making use of Shimura's canonical models and extend the result of Chen-Yui ([1] Theorem 3.7.5(2)) to the case where (a, b, N) $\neq$ N or (a/N, N) $\neq$ 1 for a positive integer N > 1.

• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.254-264
• /
• 2005

In this paper, an advanced E-Model was proposed in order to overcome disadvantages of conventional method. A new model makes the accurate VoIP call quality assessment possible by applying the burst packet loss and recency effect. In order to assess the performance of this advanced E-Model, we gained the estimated MOS value from NR(Network R) value and UR(User R) value resulted from the burst packet loss values by Gilbert Model. Through simulations and comparisons with conventional models such as MOS, PESQ, and I-Model, we reach a conclusion that advanced E-Model is more accurate and reliable method than conventional models.

• Nuclear Engineering and Technology
• /
• v.20 no.1
• /
• pp.54-64
• /
• 1988

This paper assesses the models in the SCDAP code using the results of the FLHT-2 test. Calculations show that the SCDAP correctly predicts Ire temperatures, oxidation front movement, overall hydrogen generation and peak generation rate, internal fuel rod pressure, and cladding rupture due to ballooning. A comparison of the calculated results with measured data shows that two phase level is underpredicted, and that radiation heat transfer and auto-catalytic reaction temperature of zircaloy are overpredicted. These models are recommended to be modified. The analysis also shows that the simulation of the gap in a fuel rod improves the code prediction on core damage progression.