• Journal of Korea Society of Digital Industry and Information Management
• /
• v.10 no.2
• /
• pp.29-36
• /
• 2014

Finite failure NHPP software reliability models presented in the literature exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, proposes the Gompertz distribution reliability model, which made out efficiency application for software reliability. Algorithm to estimate the parameters used to maximum likelihood estimator and bisection method, model selection based on mean square error (MSE) and coefficient of determination$(R^2)$, for the sake of efficient model, was employed. Analysis of failure using real data set for the sake of proposing fixed shape parameter of the Gompertz distribution was employed. This analysis of failure data compared with the Gompertz distribution model of shape parameter. In order to insurance for the reliability of data, Laplace trend test was employed. In this study, the proposed Gompertz model is more efficient in terms of reliability in this area. Thus, Gompertz model can also be used as an alternative model. From this paper, software developers have to consider the growth model by prior knowledge of the software to identify failure modes which can was helped.

• Journal of Ocean Engineering and Technology
• /
• v.17 no.2
• /
• pp.54-59
• /
• 2003

The characteristics of the parameters for the probability distribution of fatigue crack growth life, using the non-Gaussian random process simulation method is investigated. In this paper, the material resistance to fatigue crack growth is treated as a spatial random process, which varies randomly on the crack surface. Using the previous experimental data, the crack length equals the number of cycle curves that are simulated. The results are obtained for constant stress intensity factor range conditions with stress ratios of R=0.2, three specimen thickness of 6, 12 and 18mm, and the four stress intensity level. The probability distribution function of fatigue crack growth life seems to follow the 3-parameter Wiubull,, showing a slight dependence on specimen thickness and stress intensity level. The shape parameter, $\alpha$, does not show the dependency of thickness and stress intensity level, but the scale parameter, $\beta$, and location parameter, ${\gamma}$, are decreased by increasing the specimen thickness and stress intensity level. The slope for the stress intensity level is larger than the specimen thickness.

• The Korean Journal of Applied Statistics
• /
• v.29 no.3
• /
• pp.513-523
• /
• 2016

Fitting a mixture of multivariate skew normal distribution (MSNMix) with multiple skewness parameter vectors via EM algorithm often requires a highly expensive computational cost to calculate the moments and probabilities of multivariate truncated normal distribution in E-step. Subsequently, it is common to fit an asymmetric data set with MSNMix with a simple skewness parameter vector since it allows us to compute them in E-step in an univariate manner that guarantees a cheap computational cost. However, the adaptation of a simple skewness parameter is unrealistic in many situations. This paper proposes an approximate estimation for the MSNMix with multiple skewness parameter vectors that also allows us to treat them in an univariate manner. We additionally provide some experiments to show its effectiveness.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2002.10a
• /
• pp.301-306
• /
• 2002

The characteristics of parameters for the probability distribution of fatigue crack growth lives by the non-Gaussian random process simulation method is investigated. In this paper, the material resistance to fatigue crack growth is treated as a spatial random process, which varies randomly on the crack surface. Using the previous experimental data, the crack length - the number of cycles curves are simulated. The results are obtained for constant stress intensity factor range conditions with stress ratio of R=0.2, three specimen thickness of 6, 12 and 18mm, and the four stress intensity level. The probability distribution function of fatigue crack growth lives seems to follow the 3-parameter Wiubull and shows a slight dependence on specimen thickness and stress intensity level. The shape parameter, ${\alpha}$, does not show the dependency of thickness and stress intensity level, but the scale parameter, ${\beta}$, and location parameter, ${\upsilon}$, are decreased by increasing the specimen thickness and stress intensity level. The slope for the stress intensity level is larger than the specimen thickness.

• Communications for Statistical Applications and Methods
• /
• v.16 no.4
• /
• pp.647-658
• /
• 2009

It is already known from the previous study that flood seems to have heavier tail. Therefore, to make prediction of future extreme label, some agreement of tail behavior of extreme data is highly required. The LH-moments estimation method, the generalized form of L-moments is an useful method of characterizing the upper part of the distribution. LH-moments are based on linear combination of higher order statistics. In this study, we have formulated LH-moments of five distributions useful in hydrology such as, two types of three parameter kappa distributions, beta-${\kappa}$ distribution, beta-p distribution and a generalized Gumbel distribution. Using LH-moments reduces the undue influences that small sample may have on the estimation of large return period events.

• International Journal of Reliability and Applications
• /
• v.18 no.2
• /
• pp.83-98
• /
• 2017

Stress-strength reliability problems arise frequently in applied statistics and related fields. In the context of reliability, the stress-strength model describes the life of a component, which has a random strength X and is subjected to random stress Y. The component fails at the instant that the stress applied to it exceeds the strength and the component will function satisfactorily whenever X > Y. The problem of estimation the reliability parameter in a stress-strength model R = P[Y < X], when X and Y are two independent two-parameter Lindley random variables is considered in this paper. The maximum likelihood estimator (MLE) and Bayes estimator of R are obtained. Also, different confidence intervals of R are obtained. Simulation study is performed to compare the different proposed estimation methods. Example in real data is used as practical application of the proposed procedure.

• Communications for Statistical Applications and Methods
• /
• v.24 no.5
• /
• pp.519-531
• /
• 2017

We consider goodness-of-fit test statistics for Weibull distributions when data are randomly censored and the parameters are unknown. Koziol and Green (Biometrika, 63, 465-474, 1976) proposed the $Cram\acute{e}r$-von Mises statistic's randomly censored version for a simple hypothesis based on the Kaplan-Meier product limit of the distribution function. We apply their idea to the other statistics based on the empirical distribution function such as the Kolmogorov-Smirnov and Liao and Shimokawa (Journal of Statistical Computation and Simulation, 64, 23-48, 1999) statistics. The latter is a hybrid of the Kolmogorov-Smirnov, $Cram\acute{e}r$-von Mises, and Anderson-Darling statistics. These statistics as well as the Koziol-Green statistic are considered as test statistics for randomly censored Weibull distributions with estimated parameters. The null distributions depend on the estimation method since the test statistics are not distribution free when the parameters are estimated. Maximum likelihood estimation and the graphical plotting method with the least squares are considered for parameter estimation. A simulation study enables the Liao-Shimokawa statistic to show a relatively high power in many alternatives; however, the null distribution heavily depends on the parameter estimation. Meanwhile, the Koziol-Green statistic provides moderate power and the null distribution does not significantly change upon the parameter estimation.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.19 no.10
• /
• pp.947-952
• /
• 2006

It has been confirmed that the inner defect of insulator and the perfect diagnosis for aging are closely related to safe electric power transmission system and that the detection of accident and diagnosis technique turn out to be very important issues. But perfect diagnosis is difficult because discharge pattern is irregular. Thus, we investigated discharge pattern using the new distribution statistical models with cross-inked polyethylene(XLPE) specimens. Voltage was applied to power frequency by step method, and calibration of discharge was set to 50 pC. After the voltage was applied, it measured the discharge occurring during 10s. We investigated discharge pattern using the K-means analysis and Weibull function. We also investigated variation of centroid and shape parameter due to variation of voltage. As a result of analyzing K-means, it was confirmed that cluster including many object numbers was formed by the presence of void. And result of Weibull distribution, it was confirmed that shape parameter of discharge varied from 1.28 to 1.62 in no void specimens, and that shape parameter of discharge number varied from 1.28 to 1.62. In the void, shape parameter of discharge varied from 5.66 to 6.43, and shape parameter of discharge number varied from 5.05 to 5.08.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.2
• /
• pp.239-245
• /
• 1997

By assuming a general progressive Type-II censored sample, we propose the minimum risk estimator (MRE) and the approximate maximum likelihood estimator (AMLE) of the scale parameter of the one-parameter exponential distribution. An example is given to illustrate the methods of estimation discussed in this paper.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.6
• /
• pp.1213-1223
• /
• 2009

In this paper, we develop noninformative priors for two parameter Pareto distribution. Specially, we derive Jereys' prior, probability matching prior and reference prior for the parameter of interest. In our case, the probability matching prior is only a first order matching prior and there does not exist a second order matching prior. Some simulation reveals that the matching prior performs better to achieve the coverage probability. A real example is also considered.