• Journal of Mechanical Science and Technology
• /
• 제16권1호
• /
• pp.60-74
• /
• 2002

In this study, normal, log-normal, triangular, uniform. Weibull, Erlang and unit beta probability density functions are tried to represent the behaviour of frequency distributions of workpiece dimensions collected from various manufacturing firms. Among the distribution functions, the unit beta distribution function is found to be the best fit using the chi-square test of fit. An attempt is made for the adoption of the unit beta model to x-bar charts of quality control in manufacturing. In this direction, upper and lower control limits (UCL and LCL) of x-bar control charts of dimension measurements are estimated for the beta model, and the observed differences between the beta and normal model control limits are discussed for the measurement sets.

• Journal of Preventive Medicine and Public Health
• /
• 제20권2호
• /
• pp.261-269
• /
• 1987

In order to investigate working conditions of underground coal mines, this work was undertaken to evaluate the respirable dust and the concentration of quartz in Taeback and Kangneung areas. The concentration of quartz was determined by Fourier Transform Infrared Spectrophotometry. The results were as follows; 1) The concentration of respirable dust of drilling and coal face in Taeback and Kangneung areas were as followed; Arithmetic $Mean{\pm}S.D.(mg/m^3)$ Taeback Drilling: $2.00{\pm}1.56$ Taeback Coal Face: $3.74{\pm}3.14$ Kangneung Drilling: $4.55{\pm}4.51$ Kangneung Coal Face: $5.77{\pm}4.53$ Geometric $Mean{\pm}S.D.(mg/m^3)$ Taeback Drilling: $1.34{\pm}2.81$ Taeback Coal Face : $2.55{\pm}2.61$ Kangneung Drilling : $2.44{\pm}3.63$ Kangneung Coal Face: $4.24{\pm}2.37$ 2) Distribution of respirable dust was well fitted to the log-normal distribution and geometric mean value was $log^{-1}\;0.37{\pm}log^{-1}\;0.47(2.34{\pm}2.95)mg/m^3$. 3) The difference of respirable dust concentrations in Taeback and Kangneung areas was not significant statistically (p>0.05). 4) The concentration of quartz of drilling and coal face in Taeback and Kangneung areas were as followed; Arithmetic $Mean{\pm}S.D.(%)$ Taeback Drilling: $6.18{\pm}5.52$ Taeback Coal Face: $1.89{\pm}1.54$ Kangneung Drilling: $3.54{\pm}2.12$ Kangneung Coal Face: $2.05{\pm}3.37$ Geometric $Mean{\pm}S.D.(%)$ Taeback Drilling: $4.24{\pm}2.59$ Coal Face: $1.39{\pm}2.22$ Kangneung Drilling : $2.55{\pm}3.08$ Kangneung Coal Face : $1.24{\pm}2.33$ 5) Distribution of quartz concentrations was well fitted to the log-normal distribution and geometric mean value was $log^{-1}\;0.33{\pm}log^{-1}\;0.45(2.14{\pm}2.82)%$. 6) The difference of quartz concentrations in Taeback and Kangneung areas was not significant (p>0.05), but significant at drilling sites and coal faces (p<0.05).

• 한국대기환경학회지
• /
• 제2권3호
• /
• pp.34-44
• /
• 1986

Particle size distribution of airborne suspended particulate concentrations according to particle size in the events of yellow sand phenomena, have been measured and analyzed by using Andersen air sampler for four years, January 1982 through December 1985. The conclusions are as follows: 1. Yellow sand phenomena, generally, occur between March and May. 2. The frequent occurrences of yellow sand were observed during March and April and airborne suspended particulate concentrations in the cases of yellow sand appeared to be 2 $\sim$ 3.4 times higher than those of normal conditions. 3. Geometric mean particle diameter and its geometric mean standard deviation by logarithmic normal distribution sheet, were quite close to each other and log-distribution curves showed similar shapes. 4. Analysis by particle size distribution curve showed bi-modal distribution. 5. Concentrations of coarse particles in normal conditions were 1.2 $\sim$ 2 times higher than those of fine particles and, similarly, coarse particle concentrations in yellow sand cases were 1.3 $\sim$ 2.5 times higher than those of fine particles. 6. Concentrations of coarse particles in yellow sand cases were 2 $\sim$ 3.6 times higher than those in normal conditions and those of fine particles were 1.7 $\sim$ 3.5 times higher.

• Asian Journal of Atmospheric Environment
• /
• 제6권4호
• /
• pp.304-313
• /
• 2012

In this study, the ${\AA}$ngstrom exponent for polydispersed aerosol during dynamic processes was investigated. Log-normal aerosol size distribution was assumed, and a sensitivity analysis of the ${\AA}$ngstrom exponent with regards the coagulation and condensation process was performed. The ${\AA}$ngstrom exponent is expected to decrease because of the particle growth due to coagulation and condensation. However, it is difficult to quantify the degree of change. In order to understand quantitatively the change in the ${\AA}$ngstrom exponent during coagulation and condensation, different real and imaginary parts of the refractive index were considered. The results show that the ${\AA}$ngstrom exponent is sensitive to changes in size distribution and refractive index. The total number concentration decreases and the geometric mean diameter of aerosols increase during coagulation. On the while, the geometric standard deviation approaches monodispersed size distribution during the condensation process, and this change in size distribution affects the ${\AA}$ngstrom exponent. The degree of change in the ${\AA}$ngstrom exponent depends on the refractive index and initial size distribution, and the size parameter changes with the ${\AA}$ngstrom exponent for a given refractive index or chemical composition; this indicates that the size distribution plays an important role in determining the ${\AA}$ngstrom exponent as well as the chemical composition. Subsequently, this study shows how the ${\AA}$ngstrom exponent changes quantitatively during the aerosol dynamics processes for a log-normal aerosol size distribution for different refractive indices; the results showed good agreement with the results for simple analytic size distribution solutions.

• 대한토목학회논문집
• /
• 제31권5B호
• /
• pp.431-438
• /
• 2011

본 연구에서는 레이더 및 우량계 강우자료가 대수정규분포를 따른다고 가정하는 경우의 편의보정 문제를 살펴보았다. 이를 위해 대수정규분포를 따르는 두 강우자료의 평균, 중앙값 및 최빈값에 대한 회귀선을 유도하였으며, 레이더 강우 보정에 대한 각 회귀선의 효용성을 검토하였다. 추가로 강우자료가 정규분포를 따른다고 가정하는 경우와의 비교를 수행하였다. 본 연구는 2003년 발생한 태풍 매미의 강우자료에 대한 회귀분석 결과를 적용 예로 살펴보았다. 그 결과, 레이더 강우를 독립변수로 적용한 경우, 세 개의 회귀선을 이용한 레이더 강우의 보정결과가 모두 과소 추정된 것으로 나타났다. 반대로 우량계 강우를 독립변수로 한 경우의 세 회귀선에 대한 결과에서는 레이더 강우가 모두 과대 보정되었다. 고려한 경우 중에는 평균값에 대한 결과가 가장 우수한 것으로 나타났다. 그러나 이 경우도 정규분포로 가정하는 경우 중 우량계 강우를 독립변수로 한 원점을 통과하는 회귀선에 비해 열등한 결과를 주는 것으로 확인되었다. 따라서 이를 통해 레이더 강우의 편의 보정에 대수정규분포를 고려한 회귀식을 이용해야 한다고 결론 내리기는 힘들 것으로 보인다.

• The Korean Journal of Physiology and Pharmacology
• /
• 제13권5호
• /
• pp.367-371
• /
• 2009

The estimation of 90% parametric confidence intervals (CIs) of mean AUC and Cmax ratios in bioequivalence (BE) tests are based upon the assumption that formulation effects in log-transformed data are normally distributed. To compare the parametric CIs with those obtained from nonparametric methods we performed repeated estimation of bootstrap-resampled datasets. The AUC and Cmax values from 3 archived datasets were used. BE tests on 1,000 resampled data sets from each archived dataset were performed using SAS (Enterprise Guide Ver.3). Bootstrap nonparametric 90% CIs of formulation effects were then compared with the parametric 90% CIs of the original datasets. The 90% CIs of formulation effects estimated from the 3 archived datasets were slightly different from nonparametric 90% CIs obtained from BE tests on resampled datasets. Histograms and density curves of formulation effects obtained from resampled datasets were similar to those of normal distribution. However, in 2 of 3 resampled log (AUC) datasets, the estimates of formulation effects did not follow the Gaussian distribution. Bias-corrected and accelerated (BCa) CIs, one of the nonparametric CIs of formulation effects, shifted outside the parametric 90% CIs of the archived datasets in these 2 non-normally distributed resampled log (AUC) datasets. Currently, the 80~125% rule based upon the parametric 90% CIs is widely accepted under the assumption of normally distributed formulation effects in log-transformed data. However, nonparametric CIs may be a better choice when data do not follow this assumption.

• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.607-617
• /
• 2003

This paper offers a practical comparison of efficiency between local likelihood approach and conventional kernel approach in density estimation. The local likelihood estimation procedure maximizes a kernel smoothed log-likelihood function with respect to a polynomial approximation of the log likelihood function. We use two types of data driven bandwidths for each method and compare the mean integrated squares for several densities. Numerical results reveal that local log-linear approach with simple plug-in bandwidth shows better performance comparing to the standard kernel approach in heavy tailed distribution. For normal mixture density cases, standard kernel estimator with the bandwidth in Sheather and Jones(1991) dominates the others in moderately large sample size.

• 대한토목학회논문집
• /
• 제28권1B호
• /
• pp.55-64
• /
• 2008

기후변화에 대한 명백한 증거가 전 세계적으로 관찰되고 있음에도 불구하고 현재 사용 중인 홍수 빈도분석 방법은 이러한 기후변화나 장기변동성을 고려할 요소를 갖추지 못하고 있다. 본 연구에서는 관측된 연최대 일강우량과 일유출량 시계열을 대상으로 추세분석을 실시하여 전 지구적으로 나타난 기온상승과 같은 증가추세가 존재하는지 linear regression과 Mann-Kendall 기법을 이용하여 살펴보았으며, 나아가 기후의 변동성으로 인해 발생할 수 있는 홍수량의 증가추세를 반영한 빈도분석 방안을 제시하였다. 5개 대상지점(서울, 인천, 울릉도, 전주, 강릉)의 연최대 일강우량 모두 시간에 따른 증가추세를 일관되게 보이고 있었으나, 통계적인 유의성이 검증되지는 않았다. 홍수량도 3개의 대상지점(안동댐, 소양강댐, 대청댐) 모두에서 시간에 따른 증가추세가 관찰되었으나, 안동댐의 상향추세만이 통계적인 유의성을 내포하였다. 선형추세를 가진 홍수량의 빈도분석 및 위험도를 추정할 수 있는 대수정규 추세모형(log-normal trend model)을 소개하고, 안동댐과 소양강댐의 홍수 빈도분석을 위해 적용하였다. 적용결과 대수정규 추세모형의 2005년 50년 빈도 홍수량은 안동댐과 소양강댐 모두 대수정규 모형보다 각각 41%와 21% 증가하였으며, 목표연도가 증가함에 따라 추정되는 홍수량 역시 함께 증가함을 확인하였다.

• 대한전기학회논문지:전기물성ㆍ응용부문C
• /
• 제50권3호
• /
• pp.130-135
• /
• 2001

In this paper, the distribution characteristics of salt contaminants with the distance from sea in East coast, from Sokcho to Pusan of Korea peninsula were investigated to evaluate the design standard of KEPCO. To get the equivalent salt deposit density(ESDD), conventional brush wiping method was used. As the measuring period is comparatively short, and the measuring interval is long to check the maximum value, acquired ESDD data is very lower than the recommended value in the standard. The measured data didn't follow normal distribution, so it should take the statistical treatment. Through normalizing method, we could get a reliable probability data. In the past investigation, the accumulation characteristics of Japan is consulted to set the criterion, but the climatic condition of Korea is different to Japan. With the comparison of precipitation data and some measured data for long tern accumulation, we could set appropriate accumulation factor.

• 보험의학회지
• /
• 제3권1호
• /
• pp.103-141
• /
• 1986

The present study was undertaken to establish the decision standard of builds for the insured by using the ratio of weight-for-height as build index. Materials being examined were the ratio of weight-for-height being calculated from the actually measured heights and weights of a total of 15,838 insured persons who were examined medically at Honam Medical Department of Dong Bang Life Insurance Company, Ltd. from June, 1979 to September, 1985. The ratio of weight-for-height is calculated by the following formula. The ratio of weight-for-height(%)=$\frac{weight(kg){\times}100}{\{height(cm)-100\}{\times}0.9(kg)$ The results were as follows: 1. The distribution of the ratio of weight. for-height of the 15,838 insureds follows Log normal distribution being skewed to the left(the direction of underweight). 2. The ratio of weight-for-height were Log transformed to lead to a sym metrical pattern of distribution in which statistical rules are known to be applied more exactly. Thereafter, the establishment of dicision standard of builds was undertaken by using the log of the ratio of weight-for-height as build index. Through all ages in male, the ratio of weight-for-height indicating the range of standard lives including slight overweighted and underweighted lives besides normal lives is 80-130%, and corresponds to $"M-2{\delta}"-"M+1.5{\delta}"$ and to $M{\pm}20%$ ; in female, 85-135%, and corresponds to $"M-2{\delta}"-"M+1.5{\delta}"$ and to $M{\pm}20%$. Through all ages in male, the ratio of weight-for-height indicating the initial level of super-overweighted and super-underweighted lives is 130-150% and 75-80%,and corresponds to $M+3{\delta}\;and\;M-3{\delta}$ and to M+40% and M-25% respectively;in female, 140-160% and 75-80%, and corresponds to $M+3{\delta}\;and\;M-3{\delta}$ and to M+40%-+50% and M-25% respectively. 3. Author's rating table model for builds(a table of weight per height) is proposed. On the table, the ratings for builds, i. e. standard, super-weighted and super-underweighted lives, are listed.