• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.361-366
• /
• 2013

Prediction of process parameters is very important in parameter design. If predictions are fairly accurate, the quality improvement process will be useful to save time and reduce cost. The concept of dual response approach based on response surface methodology has widely been investigated. Dual response approach may take advantages of optimization modeling for finding optimum setting of input factor by separately modeling mean and variance responses. This study proposes an alternative dual response approach based on machine learning techniques instead of statistical analysis tools. A hybrid neural network-genetic algorithm has been proposed for the purpose of parameter design. A neural network is first constructed to model the relationship between responses and input factors. Mean and variance responses correspond to output nodes while input factors are used for input nodes. Using empirical process data, process parameters can be predicted without performing real experimentations. A genetic algorithm is then applied to find the optimum settings of input factors, where the neural network is used to evaluate the mean and variance response. A drug formulation example from pharmaceutical industry has been studied to demonstrate the procedures and applicability of the proposed approach.

• Korean Management Science Review
• /
• v.29 no.1
• /
• pp.33-56
• /
• 2012

Domestic LCD TV market is composed of two groups of products produced by major firms and small and medium enterprises. The major companies make the price relatively high, but the other makes lower in the same sizes. The model of the low price products does not make consumers choice when they choose LCD TV. This makes the questions of capability between difference price products. The reason above mentioned, the firms that include group of comparatively lower price, are worried about not increasing sale because of prejudice. This study is to find any interrelationship and evaluate the efficiency between the products using performance, exterior and brand power of product. In order to do this, a hybrid AHP/DEA evaluation model for comparison/valuation of LCD TV products is developed. The proposed process is; first, to derive hierarchy structure of LCD TV evaluation criteria using axiomatic design, second, to calculate the score of each LCD TV product through AHP analysis including weight calculation of evaluation criteria, and last, to evaluate the efficiency of LCD TV product by applying DEA by defining product scores as output and prices as input. It concludes that the high price products shows good efficiency, but there are some products with good exterior and brand power, not performance, also presenting good efficiency.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.7
• /
• pp.378-388
• /
• 2000

In this paper, we propose Fuzzy Polynomial Neural Networks(FPNN) based on Polynomial Neural Networks(PNN) and Fuzzy Neural Networks(FNN) for model identification of complex and nonlinear systems. The proposed FPNN is generated from the mutually combined structure of both FNN and PNN. The one and the other are considered as the premise part and consequence part of FPNN structure respectively. As the consequence part of FPNN, PNN is based on Group Method of Data Handling(GMDH) method and its structure is similar to Neural Networks. But the structure of PNN is not fixed like in conventional Neural Networks and self-organizing networks that can be generated. FPNN is available effectively for multi-input variables and high-order polynomial according to the combination of FNN with PNN. Accordingly it is possible to consider the nonlinearity characteristics of process and to get better output performance with superb predictive ability. As the premise part of FPNN, FNN uses both the simplified fuzzy inference as fuzzy inference method and error back-propagation algorithm as learning rule. The parameters such as parameters of membership functions, learning rates and momentum coefficients are adjusted using genetic algorithms. And we use two kinds of FNN structure according to the division method of fuzzy space of input variables. One is basic FNN structure and uses fuzzy input space divided by each separated input variable, the other is modified FNN structure and uses fuzzy input space divided by mutually combined input variables. In order to evaluate the performance of proposed models, we use the nonlinear function and traffic route choice process. The results show that the proposed FPNN can produce the model with higher accuracy and more robustness than any other method presented previously. And also performance index related to the approximation and prediction capabilities of model is evaluated and discussed.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.23 no.6
• /
• pp.622-634
• /
• 2003

This paper is concerned with a computational method for recovering a crack shape of steam generator tubes of nuclear plants. Problems on the shape identification are discussed arising in the characterization of a structural defect in a conductor using data of eddy current inspection. A surface defect on the generator tube ran be detected as a probe impedance trajectory by scanning a pancake type coil. First, a mathematical model of the inspection process is derived from the Maxwell's equation. Second, the input and output relation is given by the approximate model by virtue of the hybrid use of the finite element and boundary element method. In that model, the crack shape is characterized by the unknown coefficients of the B-spline function which approximates the crack shape geometry. Finally, a parameter estimation technique is proposed for recovering the crack shape using data from the probe coil. The computational experiments were successfully tested with the laboratory data.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.7 no.2
• /
• pp.132-137
• /
• 2007

Experimental software data capturing the essence of software projects (expressed e.g., in terms of their complexity and development time) have been a subject of intensive modeling. In this study, we introduce a new category of Hybrid Fuzzy Neural Networks (gHFNN) and discuss their comprehensive design methodology. The gHFNN architecture results from highly synergistic linkages between Fuzzy Neural Networks (FNN) and Polynomial Neural Networks (PNN). We develop a rule-based model consisting of a number of "if-then" statements whose antecedents are formed in the input space and linked with the consequents (conclusion pats) formed in the output space. In this framework, FNNs contribute to the formation of the premise part of the overall network structure of the gHFNN. The consequences of the rules are designed with the aid of genetically endowed PNNs. The experiments reported in this study deal with well-known software data such as the NASA dataset. In comparison with the previously discussed approaches, the proposed self-organizing networks are more accurate and yield significant generalization abilities.

• Environmental and Resource Economics Review
• /
• v.15 no.1
• /
• pp.27-50
• /
• 2006

We estimate $CO_2$ emissions in Korea industry, 1990 and 2000 using a commodity- by-industry IO model ($CO_2$ hybrid IO mode]). Estimated $CO_2$ emissions in industries include both $CO_2$ emissions from direct and indirect consumption. The results show that total $CO_2$ emissions has increased by 51.6 million TC (Tonne of Carbon) from 64.4 million TC in 1990 to 115.5 million TC in 2000. By applying the structural decomposition analysis technique, we decompose change of $CO_2$ emissions in Korea industry between the period 1990~2000. In the decomposition, we figure out two contributing factors, changes in $CO_2$ coefficient and changes in final demand. The latter is further decomposed as growth effects and structural effects. We also estimated each factor's contribution to the changes in $CO_2$ emissions in industries between 1990~2000. The analysis can be used as a useful resource for policy makers in improving the effectiveness of $CO_2$ emissions mitigation policy.

• Journal of Korea Technology Innovation Society
• /
• v.11 no.2
• /
• pp.264-286
• /
• 2008

In 2006, the share of energy in Korea amounted to 28% from the total import, 97% from overseas dependency, and 83% for the national Greenhouse Gas (GHG) emission in 2004. Thus, from the aspects of economical and environmental policies, an energy analysis is very important, for the industry to cope with the imminent pressure for climate change. However, the estimation of GHG gas emissions due to an energy use is still done in a primitive way, whereby each industry's usage is multiplied by coefficients recommended from international organizations in Korea. At this level, it is impossible to formulate the prevailing logic and policies in face of a new paradigm that seeks to force participation of developing countries through so called post-Kyoto Protocol. In this study, a hybrid energy input-output (E-IO) analysis is conducted on the basis of the input-output(IO) table of 2000 issued by the Bank of Korea in 2003. Furthermore, according to economic sectors, emission of the GHG relative to an energy use is characterized. The analysis is accomplished from four points of view as follows: 1) estimating the GHG emission intensity by 96 sectors, 2) measuring the contribution ratio to GHG emissions by 14 energy sources, 3) calculating the emission factor of 3 GHG compounds, and 4) estimating the total amount of national GHG emission. The total amount estimated in this study is compared with a national official statistical number. The approach could be an appropriate model for the recently spreading concept of a Life Cycle Analysis as it analyzes not only a direct GHG emission from a direct energy use but also an associated emission from an indirect use. We expect this model can provide a form for the basis of a future GHG reduction policy making.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.4
• /
• pp.679-690
• /
• 2011

Two widely used approaches for improving the quality of the output of a process are statistical process control (SPC) and automatic process control (APC). In recent hybrid processes that combine aspects of the process and parts industries, process variations due to both the inherent wandering and special causes occur commonly, and thus simultaneous application of APC and SPC schemes is needed to effectively keep such processes close to target. The simultaneous implementation of APC and SPC schemes is called integrated process control (IPC). In the IPC procedure, the output variables are monitored during the process where adjustments are repeatedly done by its controller. For monitoring the APC-controlled process, control charts can be generally applied to the output variable. However, as an alternative, some authors suggested that monitoring the input variable may improve the chance of detection. In this paper, we evaluate the performance of several monitoring statistics, such as the output variable, the input variable, and the difference variable, for efficiently monitoring the APC-controlled process when we assume IMA(1,1) noise model with a minimum mean squared error adjustment policy.

• Journal of Environmental Science International
• /
• v.9 no.1
• /
• pp.35-42
• /
• 2000

In this study, the movable bed model testing was carried out so as to analyze bed profile changes including predicting scouring and deposition of bed profile and to solve hydraulic problems affecting with bed and both-bank between upstream and downstream of intake weir in the Nakdong river channel. The movable bed model testing consists of fundamental test, movable model test and numerical analysis method respectively. The fundamental test was enforced to analyze relationship of discharge and sediment load in the tilting flume. When the movable model test was worked, it was shown that sediment budget between input sediment load and output sediment load was balanced exactly. As a result of movable model test, it was presented that scouring and deposition changes in quantities between the upstream and downstream of modification weir were less than those of nature and planning weir. Finally, numerical analysis method was operated by 1-dimensional bed profile changes model ; HEC-6 model so as to complement unsolving hard problems during movable model test. So, modification weir will sustained the stable bed profile changes than any other weirs in the study channel.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.999-1004
• /
• 2005

An algorithm for a hybrid controller consists of a sliding mode control part and a fuzzy logic part which ar purposely for nonlinear systems. The sliding mode part of the solution is based on "eigenvalue/vector"-type controller is used as the backstepping approach for tracking errors. The fuzzy logic part is a Mamdani fuzzy model. This is designed by applying sliding mode control (SMC) method to the dynamic model. The main objective is to keep the update dynamics in a stable region by used SMC. After that the plant behavior is presented to train procedure of adaptive neuro-fuzzy inference systems (ANFIS). ANFIS architecture is determined and the relevant formulation for the approach is given. Using the error (e) and rate of error (de), occur due to the difference between the desired output value (yd) and the actual output value (y) of the system. A dynamic adaptation law is proposed and proved the particularly chosen form of the adaptation strategy. Subsequently VSC creates a sliding mode in the plant behavior while the parameters of the controller are also in a sliding mode (stable trainer). This study considers the ANFIS structure with first order Sugeno model containing nine rules. Bell shaped membership functions with product inference rule are used at the fuzzification level. Finally the Mamdani fuzzy logic which is depends on adaptive neuro-fuzzy inference systems structure designed. At the transferable stage from ANFIS to Mamdani fuzzy model is adjusted for the membership function of the input value (e, de) and the actual output value (y) of the system could be changed to trapezoidal and triangular functions through tuning the parameters of the membership functions and rules base. These help adjust the contributions of both fuzzy control and variable structure control to the entire control value. The application example, control of a mass-damper system is considered. The simulation has been done using MATLAB. Three cases of the controller will be considered: for backstepping sliding-mode controller, for hybrid controller, and for adaptive backstepping sliding-mode controller. A numerical example is simulated to verify the performances of the proposed control strategy, and the simulation results show that the controller designed is more effective than the adaptive backstepping sliding mode controller.