• Journal of the Korean Data and Information Science Society
• /
• v.23 no.1
• /
• pp.53-61
• /
• 2012

There are cases when the sample size is determined based not only on the significance level but also on on the power or type II error. In this paper, we implemented the sample size and the power calculation when both the significance level and power for testing means in normal distributions and proportions in binomial distributions. The implementation is available on a web site. Alternately, we also calculate the power for a given effect size, type I error probability and sample size.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.1 no.2
• /
• pp.190-201
• /
• 1989

This experimental investigation has been conducted to determine the effects of swirling angle and flow patterns on distributions of void fraction, bubble velocity and two-phase pressure drop in a vertical straight tube. Swirling angles of $0^{\circ}$ (non swirling), $30^{\circ}$, and $45^{\circ}$ were tested with air-water two components over a range of superficial air velocities. A transparent lucite tube of 38mm in internal diameter was used for the test section. The void fraction and bubble velocities were measured by means of a optical fiber probe at the upper part of the swirler in the test section. Pressure drops which seem to be closely related with flow patterns and swirling angle were measured by a differential pressure transducer. It is shown that the probability density functions of pressure drop demonstrate peculiar features for both swirling angles and flow patterns, whereas the distributions of void fraction and bubble velocities are parabolic and flat shape in the vicinity of tube center, respectively except bubbly flow in any swirling angle cases, and the void fraction increases with increasing swirling angle around the center of tube.

• Journal of Korean Society for Quality Management
• /
• v.42 no.1
• /
• pp.33-41
• /
• 2014

Purpose: The purpose of this paper is estimate response quality of three major portal in Korea based on the response time. In addition to response time, the response time by 1Kbyte will be analysed. Methods: Data was collected from July 2010 to November 2013 using Firebug. For comparing averages, ANOVA will be used. For comparing distributions, Chisquare test and Kolmogov-Smirnov test will be used for parametric and non parametric test respectively. Results: For response quality based on response time, Daum gets the first place, Naver the second place, and Nate the third place. But the order of the response time per 1Kbyte is different. The order is Naver, Daum and Nate. Conclusion: The response quality may be estimated using various factors. Response time is the most important factor. Daum provides the shortest response time. We could say Daum provides the best response quality. But Naver provides the shortest response time per 1Kbyte. From these results, we know reducing packets is very important thing in response time.

• International journal of steel structures
• /
• v.18 no.4
• /
• pp.1397-1409
• /
• 2018

Stiffened plates with high slenderness parameters show large out-of-plane deflections, due to elastic buckling, which may occur before the plates reach their ultimate strength. From a serviceability point of view, restriction of out-of-plane deflections exceeding the fabrication tolerance is of primary importance. Compressive strength at the serviceability limit state (SLS) for slender stiffened plates under uniaxial stress was investigated through nonlinear elasto-plastic finite element analysis, considering both geometric and material nonlinearity. Both normal and high-performance steel were considered in the study. The SLS was defined based on a deflection limit and an elastic buckling strength. Probabilistic distributions of the SLS strengths were obtained through Monte Carlo simulations, in association with the response surface method. On the basis of the obtained statistical distributions, partial safety factors were proposed for SLS. Comparisons with the ultimate strength of different design codes e.g. Japanese Code, AASHTO, and Canadian Code indicate that AASHTO and Canadian Code provide significantly conservative design, while Japanese Code matches well with a 5% non-exceedance probability for compressive strength at SLS.

• Journal of the Korean Physical Society
• /
• v.80
• /
• pp.953-963
• /
• 2022

Post-neutron emission fission product mass distributions are calculated by using pre-neutron emission fission product yields (FPYs) and neutron multiplicity. A semi-empirical model is used to calculate the pre-neutron FPY, first. Then the neutron multiplicity for each fission fragment mass is used to convert the pre-neutron FPY to the post-neutron FPY. In doing so, assumptions are made for the probability for a pre-emission fission fragment with a mass number A^* to decay to a post-emission fragment with a mass number A. The resulting post-neutron FPYs are compared with the data available. The systems where the experimental data of not only the pre- and post-neutron FPY but also neutron multiplicity are available are the thermal neutron-induced fission of ²³³U, ²³⁵U and ²³⁹Pu. Thus, we applied the model calculations to these systems and compared the calculation results with those from the GEF and the data from the ENDF and the EXFOR libraries. Both the pre- and post-neutron fission product mass distributions calculated by using the semi-empirical model and the neutron multiplicity reproduce the overall features of the experimental data.

• Geomechanics and Engineering
• /
• v.23 no.3
• /
• pp.261-273
• /
• 2020

This paper presents an efficient method utilizing user-defined computer functional codes to determine the reliability of an embankment slope with spatially varying soil properties in real time. The soils' mechanical properties varied with the soil layers that had different degrees of compaction and moisture content levels. The Latin Hypercube Sampling (LHS) for the degree of compaction and Kriging simulation of moisture content variation were adopted and programmed to predict their spatial distributions, respectively, that were subsequently used to characterize the spatial distribution of the soil shear strengths. The shear strength parameters were then integrated into the Geostudio command file to determine the safety factor of the embankment slope. An explicit metamodal for the performance function, using the Kriging method, was established and coded to efficiently compute the failure probability of slope with varying moisture contents. Sensitivity analysis showed that the proposed method significantly reduced the computational time compared to Monte Carlo simulation. About 300 times LHS Geostudio computations were needed to optimize precision and efficiency in determining the failure probability. The results also revealed that an embankment slope is prone to high failure risk if the degree of compaction is low and the moisture content is high.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.700-703
• /
• 2004

A reliability analysis and design procedure based on the design of experiment (DOE) is combined with the response surface method (RSM) for numerical efficiency. The procedure established is based on a 3$^n$ full factorial DOE for numerical quadrature using explicit formula of optimum levels and weights derived for general distributions. The full factorial moment method (FFMM) shows good performance in terms of accuracy and ability to treat non-normally distributed random variables. But, the FFMM becomes very inefficient because the number of function evaluation required increases exponentially as the number of random variables considered increases. To enhance the efficiency, the response surface moment method (RSMM) is proposed. In RSMM, experiments only with high probability are conducted and the rest of data are complemented by a quadratic response surface approximation without mixed terms. The response surface is updated by conducting experiments one by one until the value of failure probability is converged. It is calculated using the Pearson system and the four statistical moments obtained from the experimental data. A measure for checking the relative importance of an experimental point is proposed and named as influence index. During the update of response surface, mixed terms can be added into the formulation.

• Journal of the Korean Society of Safety
• /
• v.29 no.6
• /
• pp.94-98
• /
• 2014

This study was intend to develop the optimal design method of suspension bridge by the reliability analysis based on minimization of life cycle cost(LCC). The reliability analysis was performed considering aleatory uncertainties included in the result of numerical analysis. The optimal design was estimated based on life-cycle cost analysis depending on the result of reliability analysis. As the effect of epistemic uncertainty, the safety index (beta), failure probability (pf) and minimum life cycle cost were random variables. The high-level distributions were generated, from which the critical percentile values were obtained for a conservative bridge design through sensitivity assessment.

• Geomechanics and Engineering
• /
• v.5 no.5
• /
• pp.485-497
• /
• 2013

The factor of safety is the most common measure of the safety margin for slopes. When the traditionally defined factor is used in kinematic approach of limit analysis, calculations can become elaborate, and iterative methods have to be used. To avoid this inconvenience, the safety factor was defined in terms of the work rates that are part of the work balance equation used in limit analysis. It was demonstrated for two simple slopes that the safety factors calculated according to the new definition fall close to those calculated using the traditional definition. Statistical analysis was carried out to find out whether, given normal distribution of the strength parameters, the distribution of the safety factor can be approximated with a well-defined probability density function. Knowing this function would make it convenient to calculate the probability of failure. The results indicated that the normal distribution could be used for low internal friction angle (up to about $16^{\circ}$) and the Johnson distribution could be used for larger angles ${\phi}$. The data limited to two simple slopes, however, does not allow assuming these distributions a priori for other slopes.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.461-473
• /
• 2014

We consider condence intervals for high quantiles of heavy-tailed distribution. The asymptotic condence intervals based on the limiting distribution of estimators are considered together with bootstrap condence intervals. We can also apply a non-parametric, parametric and semi-parametric approach to each of these two kinds of condence intervals. We considered 11 condence intervals and compared their performance in actual coverage probability and the length of condence intervals. Simulation study shows that two condence intervals (the semi-parametric asymptotic condence interval and the semi-parametric bootstrap condence interval using pivotal quantity) are relatively more stable under the criterion of actual coverage probability.