• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.33 no.1
• /
• pp.17-23
• /
• 2010

The corporate R&D(Research and Development) has a primary role of new product development and its potential is the most crucial factor to estimate corporate future value. However, its systemic inadequacies and inefficiencies, the shorten product life-cycle to satisfy customer needs, the global operations by outsourcing strategy, and the reduction of product cost, are starting to expose to R&D business processes. The three-phased execution strategy for R&D innovation is introduced to establish master plan for new R&D model. From information technology point of view, PLM(Product Life-cycle Management) is one of the business total solutions in product development area. It is not a system, but the strategic business approach that collaboratively manage the product from beginning stage to end of life in all business areas PLM functions and capabilities are usually used as references to re-design new R&D process. BPA(Business Process Assessment) and 5DP(Design Parameters) in PI6sigma developed by Samsung SDS Consulting division are introduced to establish R&D master plan and re-design process respectively. This research provides a case study for R&D process innovation. How process assessment and PMM(Process Maturity Model) can be applied in business processes, and also it explains process re-design by 5DP method.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.5
• /
• pp.543-548
• /
• 2014

Multiple-model based speech recognition framework (MMSR) has been known to be very successful in speech recognition. Since it uses multiple hidden Markov modes (HMMs) that corresponds to various noise types and signal-to-noise ratio (SNR) values, the selected acoustic model can have a close match with the test noisy speech. However, since the number of HMM sets is limited in practical use, the acoustic mismatch still remains as a problem. In this study, we experimentally determined the optimal SNR mapping between the test noisy speech and the HMM set to mitigate the mismatch between them. Improved performance was obtained by employing the SNR mapping instead of using the estimated SNR from the test noisy speech. When we applied the proposed method to the MMSR, the experimental results on the Aurora 2 database show that the relative word error rate reduction of 6.3% and 9.4% was achieved compared to a conventional MMSR and multi-condition training (MTR), respectively.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.17 no.1
• /
• pp.111-122
• /
• 2014

The provided habitat of many services from natural capital is important. But because most ecosystem services tools qualitatively evaluated biodiversity or habitat quality, this study quantitatively analyzed those aspects using the species distribution model (MaxEnt). This study used location point data of the goat(Naemorhedus caudatus), marten(Martes flavigula), leopard cat(Prionailurus bengalensis), flying squirrel(Pteromys volans aluco) and otter(Lutra lutra) from the 3rd National Ecosystem Survey. Input data utilized DEM, landcover classification maps, Forest-types map and digital topographic maps. This study generated the MaxEnt model, randomly setting 70% of the presences as training data, with the remaining 30% used as test data, and ran five cross-validated replicates for each model. The threshold indicating maximum training sensitivity plus specificity was considered as a more robust approach, so this study used it to conduct the distribution into presence(1)-absence(0) predictions and totalled up a value of 5 times for uncertainty reduction. The test data's ROC curve of endangered mammals was as follows: growing down goat(0.896), otter(0.857), flying squirrel(0.738), marten(0.725), and leopard cat(0.629). This study was divided into two groups based on habitat: the first group consisted of the goat, marten, leopard cat and flying squirrel in the forest; and the second group consisted of the otter in the river. More than 60 percent of endangered mammals' distribution probability were 56.9% in the forest and 12.7% in the river. A future study is needed to conduct other species' distribution modeling exclusive of mammals and to develop a collection method of field survey data.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.5
• /
• pp.1033-1047
• /
• 2017

Traditional explicit authentication, which requires awareness of the user's authentication process, is a burden on the user, which is one of main reasons why users tend not to employ authentication. In this paper, we try to reduce such cost by employing implicit authentication methods, such as biometrics and location based authentication methods. We define the 4-level security assurance model, where each level is mapped to an explicit authentication method. We implement our model as an Android application, where the implicit authentication methods are touch-stroke dynamics-based, face recognition based, and the location based authentication. From user experiment, we could show that the authentication cost is reduced by 14.9% compared to password authentication-only case and by 21.7% compared to the case where 6-digit PIN authentication is solely used.

• Journal of Information Technology Services
• /
• v.11 no.4
• /
• pp.233-250
• /
• 2012

In this paper, credit card delinquency means the possibility of occurring bad debt within the certain near future from the normal accounts that have no debt and the problem is to predict, on the monthly basis, the occurrence of delinquency 3 months in advance. This prediction is typical binary classification problem but suffers from the issue of data imbalance that means the instances of target class is very few. For the effective prediction of bad debt occurrence, Support Vector Machine (SVM) with kernel trick is adopted using credit card usage and payment patterns as its inputs. SVM is widely accepted in the data mining society because of its prediction accuracy and no fear of overfitting. However, it is known that SVM has the limitation in its ability to processing the large-scale data. To resolve the difficulties in applying SVM to bad debt occurrence prediction, two stage clustering is suggested as an effective data reduction method and ensembles of SVM models are also adopted to mitigate the difficulty due to data imbalance intrinsic to the target problem of this paper. In the experiments with the real world data from one of the major domestic credit card companies, the suggested approach reveals the superior prediction accuracy to the traditional data mining approaches that use neural networks, decision trees or logistics regressions. SVM ensemble model learned from T2 training set shows the best prediction results among the alternatives considered and it is noteworthy that the performance of neural networks with T2 is better than that of SVM with T1. These results prove that the suggested approach is very effective for both SVM training and the classification problem of data imbalance.

• Journal of Korea Water Resources Association
• /
• v.51 no.7
• /
• pp.575-584
• /
• 2018

An increase in heavy rainfall and floods have been observed over South Korea due to recent abnormal weather. In this perspective, the high-resolution weather forecasts have been widely used to facilitate flood management. However, these models are known to be biased due to initial conditions and topographical conditions in the process of model building. Theretofore, a bias correction scheme is largely applied for the practical use of the prediction to flood management. This study introduces a new mean field bias correction (MFBC) approach for the high-resolution numerical rainfall products, which is based on a Bayesian Kriging model to combine an interpolation technique and MFBC approach for spatial representation of the error. The results showed that the proposed method can reliably estimate the bias correction factor over ungauged area with an improvement in the reduction of errors. Moreover, it can be seen that the bias corrected rainfall forecasts could be used up to 72 hours ahead with a relatively high accuracy.

• Journal of the Society of Naval Architects of Korea
• /
• v.56 no.4
• /
• pp.373-382
• /
• 2019

This paper shows numerical results for the estimation of the propulsor efficiency of Pre-Swirl Duct for 176k bulk carrier as well as its design method. Reynolds averaged Navier-Stokes equations have been solved and the k-epsilon model applied for the turbulent closure. The propeller rotating motion is determined using a sliding mesh technique. The design process is divided into each part of Pre-Swirl Duct, duct and Pre-Swirl Stator. The design of duct was performed first because it is located further upstream than Pre-Swirl Stator. The distribution of velocity through the duct was analyzed and applied for the design of Pre-Swirl Stator. The design variables of duct include duct angle, diameter, and chord length. Diameter, chord length, equivalent angle are considered when designing the Pre-Swirl Stator. Furthermore, a variable pitch angle stator is applied for the final model of Pre-Swirl Duct. The largest reduction rate of the delivered power in model scale is 7.6%. Streamlines, axial and tangential velocities under the condition that the Pre-Swirl Duct is installed were reviewed to verify its performance.

• Journal of the Korean Society of Industry Convergence
• /
• v.22 no.1
• /
• pp.1-9
• /
• 2019

In electric agricultural machine the gearbox is used to increase torque and lower the output speed of the motor shaft. The gearbox consists of several shafts, helical gears and spur gears works in series. Optimization plays an important role in gear design as reducing the weight or volume of a gear set will increase its service life and improve the bearing capacity. In this paper the basic design parameters for gear like shaft diameter and face width are considered as the input variables. The bending stress and material volume is considered as the objective function. ANSYS was used to investigate the bending stress when the variable was changed. Artificial Neural Network (ANN) was used to obtain the mathematical model of the system based on the bending stress behaviour. The ANN was used since the output system is nonlinear. The Genetic Algorithm (GA)　technique of optimization is used to obtain the optimized values of shaft diameter and face width on the pinion based on the ANN mathematical model and the results are compared as that obtained using the traditional method. The ANN and GA were performed using MATLAB. The simulation results were shown that the proposed algorithm was successfully calculated the value of shaft diameter and face width to obtain the minimal bending stress and material volume of the gearbox.

• Composites Research
• /
• v.34 no.6
• /
• pp.337-344
• /
• 2021

In this study, the progressive failure behavior of the composite fan blade dovetail element under tensile loading is numerically investigated through finite element(FE) simulation. The accuracy of prediction by FE simulation is verified through tensile testing. The dovetail element is one of the joints for coupling the fan blade with the disk in a turbofan engine. The dovetail element is usually made of a metal material such as titanium, but the application of composite material is being studied for weight reduction reasons. However, manufacturing defects such as drop-off ply and resin pocket inevitably occur in realizing complex shapes of the fan blade made by composite materials. To investigate the effect of these manufacturing defects on the composite fan blade dovetail element, we performed numerical simulation with FE model to compare the prediction of the FE model and the tensile test results. At this time, the cohesive zone model is used to simulate the delamination behavior. Finally, we found that FE simulation results agree with test results when considering thermal residual stress and through-thickness compression enhancement effect.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.6
• /
• pp.218-226
• /
• 2020

An experimental study was conducted on baseline model updating and damage estimation techniques for the health monitoring of offshore wind turbine tripod substructures. First, a procedure for substructure health monitoring was proposed. An initial baseline model for a scaled model of a tripod substructure was established. A baseline model was updated based on the natural frequencies and the mode shapes measured in the healthy state. A training pattern was then generated using the updated baseline model, and the damage was estimated by inputting the modal parameters measured in the damaged state into the trained neural network. The baseline model could be updated reasonably using the effective fixity model. The damage tests were performed, and the damage locations could be estimated reasonably. In addition, the estimated damage severity also increased as the actual damage severity increased. On the other hand, when the damage severity was relatively small, the corresponding damage location was detected, but it was more difficult to identify than the other cases. Further studies on small damage estimation and stiffness reduction quantification will be needed before the presented method can be used effectively for the health monitoring of tripod substructures.