• Journal of Korean Society for Quality Management
• /
• v.21 no.2
• /
• pp.17-26
• /
• 1993

Process fault diagnosis is a complicated matter because quality control problems can result from a variety of causes. These causes include problems with electrical components, mechanical components, human errors, job justification errors, and air conditioning influences. In order to make the system run smoothly with minimum delay, it is necessary to suggest heuristic remedies for the detected faults. Hence, this paper describes a heuristic methodology of fault diagnosis that is performed using statistical patterns generated by quality characteristics The proposed methodology is described briefly as follows: If a sample pattern generated by random variables is similar to the number of prototype patterns, the sample pattern may be matched by any prototype pattern among them to be resembled. This concept is based on the similarity between a sample pattern and the matched prototype pattern. The similarity is calculated as the weighted average of squared deviation, which is expressed as the difference between the relative values of standard normal distribution to be transformed by the observed values of quality characteristics in a sample pattern and the critical values of the corresponding ones in a matched prototype pattern.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.06a
• /
• pp.1108-1113
• /
• 2000

A condition monitoring methodology for rotating machinery is proposed based on multivariate statistical analysis. The CMS usually are using the vibration signal amplitude such as acceleration RMS, peak and velocity RMS to detect machine faults but the information is not so enough that CMS cannot perform reliable monitoring. So new parameters are added such as shape factor, crest factor, kurtosis and skewness as time domain parameters and spectrum amplitude of rotating frequency, $2^{nd}$ harmonics and gear mesh frequency etc. as frequency domain parameters. Many parameters are combined to represent the machine state using the Hotelling's $T^2$ statistics. The proposed methodology is tested in laboratory and the on-line experiment has shown that the proposed methodology offers a reliable monitoring for rotating machinery.

• Proceedings of the Safety Management and Science Conference
• /
• 2008.04a
• /
• pp.247-258
• /
• 2008

This paper purports to investigate the statistical basic principles through "TRIZ-used 6-sigma integrated activity design" and further, to raise "the importance of measuring technical innovative activities based on 6-sigma integrated design methodology by organization". This paper intends to investigate how the activities of the best conditions, which are necessary for the stage of 6-sigma design, are chosen in the TPM-based 6-sigma integration activities. And it will examine the options, when used together with other technique, in the probability statistical methodology. Through the combination of the situations which are chosen when activity techniques, which are mostly used for production and technology, and 6-sigma integration activities are used together, it intends to practically use as the basic data for company standards.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.50 no.6
• /
• pp.27-35
• /
• 2008

This study aims to estimate the optimal probability distribution of daily electricity generation by wind power, in order to contribute in rural green-village planning. Wind power generation is now being recognized as one of the most popular sources for renewable resources over the country. Although it is also being adapted to rural area for may reasons, it is important to estimate the magnitudes of power outputs with reliable statistical methodologies while applying historical daily wind data, for correct feasibility analysis. In this study, one of the well-known statistical methodology is employed to define the appropriate statistical distributions for monthly power outputs for specific rural areas. The results imply that the assumption of normal distributions for many cases may lead to incorrect decision-making and therefore lead to the unreliable feasibility analysis. Subjective methodology for testing goodness of fit for normal distributions on all the cases in this study, provides possibilities to consider the other various types of statistical distributions for more precise feasibility analysis.

• Journal of Microbiology and Biotechnology
• /
• v.28 no.6
• /
• pp.902-908
• /
• 2018

Optimization of the culture medium to maximize menaquinone-7 (MK-7) production by Bacillus subtilis strain KCTC 12392BP in static culture was carried out using statistical experimental methods, including one factor at a time, fractional factorial design, and response surface methodology (RSM). Maltose (carbon source), tryptone (nitrogen source), and glycerol (activator) were identified as the key medium components for MK-7 synthesis by the fractional factorial design, and were selected for statistical optimization by RSM. The statistical analysis indicated that, in the range that was studied, maltose, tryptone, and glycerol were all critical factors having profound effects on the production of MK-7, with their coefficients for linear and quadratic all significant at the p < 0.05 level. The established model was efficient and feasible, with a determination coefficient ($R^2$) of 0.9419. The predicted concentrations of maltose, tryptone, and glycerol in the optimal medium were determined as 36.78, 62.76, and 58.90 g/l, respectively. In this optimized medium, the maximum yield of MK-7 reached a remarkably high level of $71.95{\pm}1.00{\mu}g/ml$ after 9 days of static fermentation, which further verified the practicability of this optimized strategy.

• Journal of Korean Institute of Industrial Engineers
• /
• v.7 no.2
• /
• pp.55-59
• /
• 1981

An Application of Experimental Designs to Improve the quality of Industrial product : optimization Methodology of statistical model. The primary object of this paper is to aid scientists and Engineers, in applying response surface procedures to obtain operating conditions for many technical fields, particularly for industrial manufacturing processes. The problem considered in this paper is to select technically and scientifically some important factors affecting the quality of products through the experimental design and analysis of response surface. Even though the mathematical model is unknown these statistical analysis can be applicable to control the quality of industrial products and to determine optimum operating conditions for many technical fields, particularly, for industrial manufacturing processes. This paper proposes a method to obtain the optimum operating condition, and how to find the condition by using table of orthogonal array experiments, and optimization methodology of statistical model.

• 한국석유지질학회:학술대회논문집
• /
• spring
• /
• pp.10-14
• /
• 1998

A systematic methodology is developed for the electrofacies determination from wireline log data using multivariate statistical analysis. To consider corresponding contribution of each log and reduce the computational dimension, multivariate logs are transformed into a single variable through principal components analysis. Resultant principal components logs are segmented using the statistical zonation method to enhance the efficiency and quality of the interpreted results. Hierarchical cluster analysis is then used to group the segments into electrofacies. Optimal number of groups is determined on the basis of the ratio of within-group variance to total variance and core data. This technique is applied to the wells in the Korea Continental Shelf. The results of field application demonstrate that the prediction of lithology based on the electrofacies classification matches well to the core and the cutting data with high reliability This methodology for electrofacies classification can be used to define the reservoir characteristics which are helpful to the reservoir management.

• Journal of Industrial Technology
• /
• v.33 no.A
• /
• pp.41-48
• /
• 2013

A statistical downscaling methodology has been developed to investigate future daily wind speeds over South Korea. This methodology includes calibration of the statistical downscaling model by using large-scale atmospheric variables encompassing NCEP/NCAR reanalysis data, validation of the model for the calibration period, and estimation of the future wind speed based on the general circulation model (GCM) outputs of scenario A1B of the CGCM3. Based on the scenario A1B of the CGCM3 model, the potential impacts of climate change on the daily surface wind speed is relatively small (+/- 1m/s) in South Korea.

• Communications for Statistical Applications and Methods
• /
• v.7 no.3
• /
• pp.945-953
• /
• 2000

Since early 1960s, when the telephone survey was used in the research area for he first tie, there has been existed various methods to gather the information by survey. The existing survey methodology called PAPI(Paper-And-Pen Interveiw), due to the appearance of Personal Computer, might well be developed progressively. Mid-1980s, Internet was advanced remarkably in terms of technology. from early 1990s, in addition it served as a stepping-stone for progressive collecting method. Internet Survey is now called WWW Survey and expected that it will substitute for most surveys from now on. We explain the role and the characteristics for Internet Survey as one of he various data collecting methods. Furthermore, we draw the futures about questionnaires, data collecting and statistical analysis with it.

• Communications for Statistical Applications and Methods
• /
• v.24 no.6
• /
• pp.543-559
• /
• 2017

Tree-based regression and classification ensembles form a standard part of the data-science toolkit. Many commonly used methods take an algorithmic view, proposing greedy methods for constructing decision trees; examples include the classification and regression trees algorithm, boosted decision trees, and random forests. Recent history has seen a surge of interest in Bayesian techniques for constructing decision tree ensembles, with these methods frequently outperforming their algorithmic counterparts. The goal of this article is to survey the landscape surrounding Bayesian decision tree methods, and to discuss recent modeling and computational developments. We provide connections between Bayesian tree-based methods and existing machine learning techniques, and outline several recent theoretical developments establishing frequentist consistency and rates of convergence for the posterior distribution. The methodology we present is applicable for a wide variety of statistical tasks including regression, classification, modeling of count data, and many others. We illustrate the methodology on both simulated and real datasets.