• Journal of the Korea Safety Management & Science
• /
• v.7 no.3
• /
• pp.47-61
• /
• 2005

Many researchers have proved that risk measurement of information systems is a very effective tool for improving confidence of information systems. However, information system risk in Korea still includes many subjective judgements. This study deals with applying a quantitative model to improve risk measurement of information system quality. First of all, we have come up with solutions to improve the evaluation efficiency on risk measurement. We have merged the risk guidelines of COBIT and CMM, and developed a quantified evaluation scheme that call by risk point. We have proved the validity of this model by interviews with experts and by case studies.

• 한국방재학회:학술대회논문집
• /
• 2008.02a
• /
• pp.783-786
• /
• 2008

The main goal of this study is to understand the effects of direct-runoff of chemistry and organic fertilizers which are adsorbing to sediment from farmland and non-point source discharge characteristics which are discharged to stream with soil erosion when rainfalls. pollutographs of TSS, BOD, COD, TN, and TP were measured for 10 rainfall events at watershed. EMC (Event Mean Concentration) were calculated for each rainfall event using quality and quantity measured. The result shows that the EMC ranges of 95% confidence intervals are 50.5-203 mg/L for TSS, 0.8-14.2 mg/L for $BOD_5$, 4.2-20.7 mg/L for $COD_{Mn}$, 0.2-0.5 mg/L for TP, 2.4-4.5 mg/L for TN, 1.36-3.04 mg/L for NO3--N, 0.13-0.42 mg/L for NH4+-N and 0.82-1.77 mg/L for TKN.

• Journal of Applied Reliability
• /
• v.14 no.1
• /
• pp.1-9
• /
• 2014

One important measure of performance for a repairable system is steady-state availability. In this paper, a method to estimate and establish confidence interval for the steady-state availability under Weibull lifetime and lognormal repair time distributions is proposed. Also, bias and mean squared error of a point estimator for an availability are investigated. In addition, a procedure to derive the sample size and critical value for availability demonstration test is presented and illustrated with a numerical example.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 2006.11a
• /
• pp.199-204
• /
• 2006

ANOVA is widely used for measurement system analysis. It assumes that the measurement error is normally distributed, which may not be seen in some industrial cases. In this study, the estimates of the measurement system variability and PTR (precision-to-tolerance ratio) are obtained by using weighted standard deviation for the case where the measurement error is non-normally distributed. The Standard Bootstrap method is used for estimating confidence intervals of measurement system variability and PTR. The point and confidence interval estimates for the cases with normally distributed measurement error are compared to those with non-normally distributed measurement errors through computer simulation.

• Communications for Statistical Applications and Methods
• /
• v.24 no.1
• /
• pp.29-41
• /
• 2017

The estimation problem of expected time to failure of units is studied in a discrete set up. A simple step-stress accelerated life testing is considered with a Type-I censored sample from geometric distribution that is a commonly used distribution to model the lifetime of a device in discrete case. Maximum likelihood estimators as well as the associated distributions are derived. Exact, approximate and bootstrap approaches construct confidence intervals that are compared via a simulation study. Optimal confidence intervals are suggested in view of the expected width and coverage probability criteria. An illustrative example is also presented to explain the results of the paper. Finally, some conclusions are stated.

• Tunnel and Underground Space
• /
• v.5 no.2
• /
• pp.89-94
• /
• 1995

Regression analysis and a comparative study were carried out for 52 blast vibration data which were monitored by changing the monitoring distance and charge per delay. The results are as follows: 1) The square and cube root scalings and general equation which have a confidence level at the point of 90% and 99% are V90=33300(SD)-2.026 , V90=23600(SD)-1.993, V90=26300W0.755 R-2.007 and V99=48400(SD)-2.026, V99=34000(SD)-1.993 , V99=38100W0.755R-2.007, respectively. 2) There is need to decide the allowable max. charge weight per delay considering the cross points comparatively of the nomogram constructed using several predictived equations. 3) It is necessary to derive the predictive equation on the basis of blast vibration level monitored in field and to decide safe vibration level and the confidence level.

• KDI Journal of Economic Policy
• /
• v.38 no.2
• /
• pp.1-20
• /
• 2016

In this paper, we conduct uniform inference of two widely used versions of the Phillips curve, specifically the random-walk Phillips curve and the New-Keynesian Phillips curve (NKPC). For both specifications, we propose a potentially time-varying natural unemployment (NAIRU) to address the uncertainty surrounding the inflation-unemployment trade-off. The inference is conducted through the construction of what is known as the uniform confidence band (UCB). The proposed methodology is then applied to point-ahead inflation forecasting for the Korean economy. This paper finds that the forecasts can benefit from conducting UCB-based inference and that the inference results have important policy implications.

• Journal of Korean Society for Quality Management
• /
• v.35 no.1
• /
• pp.10-19
• /
• 2007

ANOVA is widely, used for measurement system analysis. It assumes that the measurement error is normally distributed, which nay not be seen in some industrial cases. In this study the estimates of the measurement system variability and PTR (precision-to-tolerance ratio) are obtained by using weighted standard deviation for the case where the measurement error is non-normally distributed. The Standard Bootstrap method is used foy estimating confidence intervals of measurement system variability and PTR. The point and confidence interval estimates for the cases with normally distributed measurement error are compared to those with non-normally distributed measurement errors through computer simulation.

• Korean Journal of Organic Agriculture
• /
• v.26 no.1
• /
• pp.33-41
• /
• 2018

The paper aims to estimate efficiency of watermelon by using a bootstrapping approach to generating efficiency estimates through Monte Carlo simulation resampling process. We use the input-output data for watermelon 107 farmers. The main results are as follows. The estimates of efficiency depends on the methodology. The estimates of general DEA is greater than the bootstrapping method. The technical efficiency and pure technical efficiency measure of watermelon is 0.72, 0.82 respectively. However the bias-corrected estimates are less than those of DEA. We know that the DEA estimator is an upward biased estimator. According to these results, the DEA bootstrapping model used here provides bias-corrected and confidence intervals for the point estimates, it is more preferable.

• International Journal of Reliability and Applications
• /
• v.2 no.1
• /
• pp.1-26
• /
• 2001

The purpose of this paper is to introduce a proportional reversed hazard rate model, in contrast to the celebrated proportional hazard model, and study some of its structural properties. Some criteria of ageing are presented and the inheritance of the ageing notions (of the base line distribution) by the proposed model are studied. Two important data sets are analyzed: one uncensored and the other having some censored observations. In both cases, the confidence bands for the failure rate and survival function are investigated. In one case the failure rate is bathtub shaped and in the other it is upside bath tub shaped and thus the failure rates are non-monotonic even though the baseline failure rate is monotonic. In addition, the estimates of the turning points of the failure rates are provided.