• Journal of the Korean Statistical Society
• /
• v.26 no.4
• /
• pp.477-491
• /
• 1997

A new predictive classification rule for assigning future cases into one of two multivariate normal population (with unknown normal mixture model) is considered. The development involves calculation of posterior probability of each possible normal-mixture model via a default Bayesian test criterion, called intrinsic Bayes factor, and suggests predictive distribution for future cases to be classified that accounts for model uncertainty by weighting the effect of each model by its posterior probabiliy. In this paper, our interest is focused on constructing the classification rule that takes care of uncertainty about the types of covariance matrices (homogeneity/heterogeneity) involved in the model. For the constructed rule, a Monte Carlo simulation study demonstrates routine application and notes benefits over traditional predictive calssification rule by Geisser (1982).

• Structural Engineering and Mechanics
• /
• v.62 no.6
• /
• pp.703-708
• /
• 2017

The goal of this study is to evaluate behavior uncertainties of structures by using interval finite element analysis based on interval factor method as a specific non-stochastic tool. The interval finite element method, i.e., interval FEM, is a finite element method that uses interval parameters in situations where it is not possible to get reliable probabilistic characteristics of the structure. The present method solves the uncertainty problems of a 2D solid structure, in which structural characteristics are assumed to be represented as interval parameters. An interval analysis method using interval factors is applied to obtain the solution. Numerical applications verify the intuitive effectiveness of the present method to investigate structural uncertainties such as displacement and stress without the application of probability theory.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.2
• /
• pp.87-93
• /
• 2008

In this paper, we propose a hierarchical fuzzy based nutrition evaluation system that can analyze the individuals' nutrition status through the inference results generated by each layer. Moreover, a method to minimize the uncertainty of inference in the evaluated nutrition status is discussed. To show the effect of the uncertainty in fuzzy inference, we compared the results of nutrition evaluation with/without the certainty factor of rules on 132 people over the age of 65. From the experimental results, we can see that the evaluation method with the modified certainty factor provides better reliability than that of the general evaluation method without the certainty factor.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.13 no.1
• /
• pp.396-404
• /
• 2021

It is important to verify that the contracted speed-power performance of a ship is satisfied in sea trials. International Organization for Standardization (ISO) has published the procedure for measuring and assessing ship speed during sea trials. The results obtained from actual sea conditions inevitably include various uncertainty factors. In this study, double run tests were performed on one container ship to analyze the uncertainty of sea trial on three maximum continuous rating conditions. The uncertainty factors and scale of uncertainty were examined based on the measured raw data during sea trial. The results indicate that the expanded uncertainty for ideal power performance is approximately ±1.4% at 95% confidence level (coverage factor k = 2) and most of the uncertainty factors were because of the shaft power measurement system.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2009.10a
• /
• pp.553-554
• /
• 2009

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.39 no.12
• /
• pp.1107-1114
• /
• 2011

In this paper, the fatigue life of a laser peened structure was predicted. In order to calculate residual stress induced by laser peening finite element simulation was carried out. Modified Goodman equation was used to consider the effect of compressive residual stress induced by laser peening in fatigue analysis. In addition, additive adjustment factor approach was applied to consider S-N curve model uncertainty. Consequently, the reliable bounds of the predicted fatigue life of the laser peened structure was determined.

• Journal of the Korean Society of Tobacco Science
• /
• v.26 no.2 s.52
• /
• pp.168-178
• /
• 2004

Uncertainty of final measurement results considering main uncertainty sources being in nicotine of mainstream smoke was estimated. This study was accomplished by using the ISO 'The Guide to the Expression of Uncertainty in Measurement'. Using the two point re-calibration method, uncertainty for nicotine concentration was calculated considering the uncertainty sources of each step. The concentration and uncertainty of nicotine in mainstream smoke was estimated as $153.95{\pm}17.84\;{\mu}g/mL\;(0.77\pm0.089 mg/cig)$. The expanded uncertainty was $17.84 {\mu}g/mL(\pm0.089 mg/cig).$ The reported expanded uncertainty of the measurement is stated as the standard uncertainty of measurement multiplied by a coverage factor of 2, which for a normal distribution corresponds to a coverage probability of approximately $95\%$ The former expression indicates the conversion concentration into the sample.

• Progress in Superconductivity
• /
• v.13 no.2
• /
• pp.97-104
• /
• 2011

In the dielectric resonator method, which has been widely used for measuring the microwave surface resistance of superconductors, accuracies in the geometrical factors (G-factors) affect the uncertainty in the measured surface resistance. We compare the G-factors of short-ended sapphire resonator as obtained by using field analysis with those by using computer simulations: The former is obtained by using the analytic expressions for the electric and the magnetic field components inside the resonator, and the latter by using computer software. The G-factors as obtained by using the latter appear to be closer to those obtained by using the former as the resonator space is divided into larger number of sub-space, i.e., a tighter mesh, with a difference of ~8 % observed for a mesh of 14400 sub-spaces reduced to ~2 % for 114996 sub-spaces. Variations in the relative uncertainty in the surface resistance of typical $YBa_2Cu_3O_{7-\delta}$ superconductor films with those in the G-factors are studied, which provides an upper limit of the relative uncertainty in the G-factors required for realizing the target uncertainty in the surface resistance. These results could be useful in estimating the optimum number of meshes for obtaining the G-factors through computer simulations.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.1
• /
• pp.31-36
• /
• 2015

There are many uncertainty items when we're working on a software. Especially, if we don't have experience in similar field, ambiguity items have a strong influence on the system entirely. Management of ambiguity items such like uncertainty things is important the factor for reliability of software. Therefore, this research is processing the evaluation criteria for remove of uncertainty items by identify of ambiguity items. Also, this research provides criteria of uncertainty identify and processing of uncertainty items, quantitative evaluation criteria to the remove of ambiguity on software development.

• Journal of the Korean Society of Tobacco Science
• /
• v.22 no.1
• /
• pp.91-98
• /
• 2000

This study was carried out to evaluate the uncertainty in the analysis of menthol content from the mentholated cigarette. Menthol in the sample cigarette was extracted with methanol containing an anethole as an internal standard, and then analyzed by gas chromatography. As the sources of uncertainty associated with the analysis of menthol, were the following points tested, such as the weighing of sample, the preparation of extracting solution, the pipetting of extracting solution into the sample, the preparation of standard solution, the precision of GC injections for standard & sample solution, the GC response factor of standard solution, the reproducibility of menthol analysis, and the determination of water content in tobacco, etc. For calculating the uncertainties, type A of uncertainty was evaluated by the statistical analysis of a series of observation, and type B by the information based on supplier's catalogue and/or certificated of calibration. Sources of uncertainty were subsequently included and mathematically combined with the uncertainty arising from the assessment of accuracy to provide the overall uncertainty. It was shown that the main source of uncertainty came from the errors in the reproducibility of menthol and water determination, the purity of menthol reference material in the preparation of standard solution, and the precision of GC injections for sample solution. The errors in sample weighing and volume measurement contributed relatively little to the overall uncertainty. The expanded uncertainty in the mentholated cigarettes, Korean and American brand, at 0.95 level of statistical confidence was $\pm$0.06 and $\pm$0.07 mg/g for a menthol content of 1.89 and 2.32 mg/g, respectively.