• Proceedings of the KIEE Conference
• /
• 1993.07a
• /
• pp.254-257
• /
• 1993

This paper presents a new method to implement Hebbian learning method on artificial neural network. In hebbian learning algorithm, complexity in terms of multiplications is high. To save the chip area, we consider a new learning circuit. By calculating similarity, or correlation between $X_i$ and $O_i$, large portion of circuits commonly used in conventional neural networks is not necessary for this new hebbian learning circuit named COR. The output signals of COR is applied to weight storage capacitors for direct control the voltages of the capacitors. The weighted sum, ${\Sigma}W_{ij}O_j$, is realized by multipliers, whose output currents are summed up in one line which goes to learning circuit or output circuit. The drain current of the multiplier can produce positive or negative synaptic weights. The pass transistor selects eight learning mode or recall mode. The layout of an learnable six-neuron fully connected Hopfield neural network is designed, and is simulated using PSPICE. The network memorizes, and retrieves the patterns correctly under the existence of minor noises.

• Journal of Korean Society for Quality Management
• /
• v.34 no.2
• /
• pp.129-135
• /
• 2006

In the advance of computer technology, it is possible to keep all the related informations for monitoring equipments in control and huge amount of real time manufacturing data in a data base. Thus, the statistical analysis of large data sets with hundreds of thousands observations and hundred of independent variables whose some of values are missing at many observations is needed even though it is a formidable computational task. A tree structured approach to classification is capable of screening important independent variables and their interactions. In a Six Sigma project handling large amount of manufacturing data, one of the goals is to screen vital few variables among trivial many variables. In this paper we have reviewed and summarized CART, C4.5 and CHAID algorithms and proposed a simple method of screening vital few variables by selecting common variables screened by all the three algorithms. Also how to develop a logistics regression model on a large data set is discussed and illustrated through a large finance data set collected by a credit bureau for th purpose of predicting the bankruptcy of the company.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.2
• /
• pp.212-219
• /
• 2012

This study focus on what methods of MOT are needed and utilized to solve a technical problem in the MOT leading company. We first classified the MOT methods by project life cycle and R&D job quality based on the several discussion with working group. In order to study, the survey was carried out by 168 experts working on MOT practices. Our main research findings are as follows. The level of utilized MOT methods was above average because MOT sector was beginning to represent an increasing share of total R&D business. But the satisfaction level on the MOT majors was below average. The more utilized MOT methods in practices were 'environment analysis and opportunity research,' 'business feasibility,' 'roadmapping,' 'portfolio,' 'technology tree,' 'scheduling,' 'risk management,' 'six sigma,' 'design of experiment,' 'quality control,' 'cost analysis' etc. And the subjects of product realization process were also needed to MOT practice such as 'design for reliability,' 'design for cost,' 'design for performance,' 'design for safety' deeply involved to product quality. Finally, the capability requested to University students were 'problem define and solving,' 'technology planning and strategy,' 'creative thinking.'

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2011.04a
• /
• pp.735-741
• /
• 2011

In the early phase of vehicle development, CAE is conducted as tool for vehicle performance assessment. To maintain acceptable road noise performance, solution for reduced vehicle sensitivity is required. Chassis interface dynamic stiffness characteristics are key component to isolating vibration and noise of road from the vehicle interior. This research provide how to set up the optimized dynamic characteristics under noise effect through DFSS study. CAE-based DOE is performed to build prediction math model, CMS process involves DOE to achieve very fast run times while giving results very comparable. Minimized $95^{th}$ percentile of performance distribution is applied to minimize vehicle sensitivity and road noise levels variation during the optimization process. Finally, the results of optimization were reviewed for performance and robustness.

• Journal of Auto-vehicle Safety Association
• /
• v.7 no.1
• /
• pp.27-32
• /
• 2015

In order to protect occupant during car crash accident, Regulation and NCAP(New Car assessment Program) have been developed among various countries like U.S.A., Europe, Korea and Australia. Especially NCAP scores affect to sales of vehicles. So vehicle makers are trying to get good score in NCAP. Low leg injuries play an important role in Australia and Euro NCAP and these injuries are related with footrest design. Optimization of footrest design in early stage of vehicle development is necessary to obtain better and robust results of low legs during crash tests. In this paper, DFSS method and finite element model were used to optimize the low leg performance in small RHD vehicles. Compared with the lower leg injury of base model, the lower leg injury of proposed model was slightly improved and robustness was enhanced also.

• Journal of ILASS-Korea
• /
• v.17 no.3
• /
• pp.140-145
• /
• 2012

Recent common-rail injector of a diesel engine needs more smaller nozzle hole to meet the stringent emission regulation. But, small nozzle hole diameter can cause nozzle coking which is occurred due to the deposits of post-combustion products. Nozzle coking has a negative effect on the performance of fuel injector because it obstructs the fuel flow inside a nozzle hole. In this study DFSS (Design for six sigma) method was applied to find the effect of nozzle design parameter on nozzle coking. Total 9 injector samples were chosen and tested at diesel engine. The results show that nozzle hole diameter and K-factor have more effect on nozzle coking than A-mass and hole length. Large hole diameter and A-mass, small hole length and K-factor give more positive performance on nozzle coking in these experimental conditions. But, a performance about nozzle coking and exhaust gas emission shows the opposite tendency. Further study is needed to find the relation between nozzle coking and emission characteristic for the optimization of injector nozzle design.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.21 no.7
• /
• pp.674-681
• /
• 2011

In the early phase of vehicle development, CAE is conducted as tool for vehicle performance assessment. To maintain acceptable road noise performance, solution for reduced vehicle sensitivity is required. Chassis interface dynamic stiffness characteristics are key component to isolating vibration and noise of road from the vehicle interior. This research provide how to set up the optimized dynamic characteristics under noise effect through DFSS study. CAE-based DOE is performed to build prediction math model, CMS process involves DOE to achieve very fast run times while giving results very comparable. Minimized 95th percentile of performance distribution is applied to minimize vehicle sensitivity and road noise levels variation during the optimization process. Finally, the results of optimization were reviewed for performance and robustness.

• Bulletin of the Korean Chemical Society
• /
• v.11 no.1
• /
• pp.49-54
• /
• 1990

The gas-phase thermolysis reactions of ${\alpha}$- and ${\beta}$-methylated ethyl formates, Y = $CH-X-CHR_1CH_2R_2$ where X = Y = O or S and $R_1\;=\;R_2$ = H or $CH_3$, are investigated theoretically using the AM1 method. The experimental reactivity order is reproduced correctly by AM1 in all cases. The thermolysis proceeds through a six-membered cyclic transition state conforming to a retro-ene reaction, which can be conveniently interpreted using the frontier orbital theory of three-species interactions. The methyl group substituted at $C_{\alpha}\;or\;C_{\beta}$ is shown to elevate the ${\pi}$-HOMO of the donor fragment (Y = C) and depress the ${\sigma}^{\ast}$-LUMO of the acceptor fragment ($C_{\beta}$-H), increasing the nucleophilicity of Y toward ${\beta}$-hydrogen which in turn increases the reactivity. The two bond breaking processes of the $C_{\alpha}$-X and $C_{\beta}$-H bonds are concerted but not synchronous so that the reaction takes place in two stages as Taylor suggested. The initial cleavage of $C_{\alpha}$-X is of little importance but the subsequent scission of $C_{\beta}$-H occurs in a rate determining stage.

• Ocean Systems Engineering
• /
• v.9 no.3
• /
• pp.329-347
• /
• 2019

The shipbuilding industry is characterized by order production, and various processes are performed simultaneously in the construction of ships. Therefore, effective management of the production process and productivity improvement form important key factors in the industry. For decades, researchers and process managers have attempted to improve processes by using business process analysis (BPA). However, conventional BPA is time-consuming, expensive, and mainly based on subjective results generated by employees, which may not always correspond to the actual conditions. This paper proposes a method to improve the production process of offshore plant modules by analysing the process mining data obtained from the shipbuilding industry. Process mining uses information accumulated from the system-provided event logs to generate a process model and determine the values hidden within the process. The discovered process is visualized as a process model. Subsequently, alternatives are proposed by brainstorming problems (such as bottlenecks or idle time) in the process. The results of this study can aid in productivity improvement (idle time or bottleneck reduction in the production process) in conjunction with a six-sigma technique or ERP system. In future, it is necessary to study the standardization of the module production processes and development of the process monitoring system.

• International Journal of Automotive Technology
• /
• v.8 no.3
• /
• pp.259-268
• /
• 2007

A knock detection circuit that is based on the signal of an accelerometer installed on the engine block of a spark ignition automotive engine has a band-pass filter with a certain frequency as a parameter to be calibrated. A new statistical method for the determination of the frequency which is the most suitable for the knock detection in real-time applications is proposed. The method uses both the cylinder pressure and block vibration signals and is divided into two steps. In both steps, a new recursive trigonometric interpolation method that calculates the frequency contents of the signals is applied. The new trigonometric interpolation method developed in this paper improves the performance of the Discrete Fourier Transformation, allowing a flexible choice of the size of the moving window. In the first step, the frequency contents of the cylinder pressure signal are calculated. The knock is detected in the cylinder of the engine cycle for which at least one value of the maximal amplitudes calculated via the trigonometric interpolation method exceeds a threshold value indicating a considerable amount of oscillations in the pressure signal; this cycle is selected as a knocking cycle. In the second step, the frequency analysis is performed on the block vibration signal for the cycles selected in the previous step. The knock detectability, which is an individual cylinder attribute at a certain frequency, is verified via a statistical hypothesis test for testing the equality of two mean values, i.e. mean values of the amplitudes for knocking and non-knocking cycles. Signal-to-noise ratio is associated in this paper with the value of t-statistic. The frequency with the largest signal-to-noise ratio (the value of t-statistic) is chosen for implementation in the engine knock detection circuit.