• The Korean Journal of Applied Statistics
• /
• v.33 no.4
• /
• pp.379-393
• /
• 2020

A terminal event such as death may censor an intermediate event such as relapse, but not vice versa in semi-competing risks data, which is often seen in medicine, public health, and epidemiology. We propose a Weibull regression model with a normal frailty to analyze semi-competing risks data when all three transition times of the illness-death model are possibly interval-censored. We construct the conditional likelihood separately depending on the types of subjects: still alive with or without the intermediate event, dead with or without the intermediate event, and dead with the intermediate event missing. Optimal parameter estimates are obtained from the iterative quasi-Newton algorithm after the marginalization of the full likelihood using the adaptive importance sampling. We illustrate the proposed method with extensive simulation studies and PAQUID (Personnes Agées Quid) data.

• The KIPS Transactions:PartD
• /
• v.8D no.4
• /
• pp.393-400
• /
• 2001

This paper presents a stochastic model for the software failure phenomenon based on a nonhomogeneous Poisson process (NHPP). The failure process is analyzed to develop a suitable mean value function for the NHPP ; expressions are given for several performance measure. Actual software failure data are compared with generalized model by Goel dependent on the constant reflecting the quality of testing. The performance measures and parametric inferences of the new models, Rayleigh and Gumbel distributions, are discussed. The results of the new models are applied to real software failure data and compared with Goel-Okumoto and Yamada, Ohba and Osaki models. Tools of parameter inference was used method of the maximun likelihood estimate and the bisection algorithm for the computing nonlinear root. In this paper, using the sum of the squared errors, model selection was employed. The numerical example by NTDS data was illustrated.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.7 no.6
• /
• pp.214-221
• /
• 1999

Spot welding is very important and useful technology in fabriaction of the thin sheet structure such as the automobile, train and air craft, Because fatigue strength of the spot welding point is however considerably lower than base metal due to stress concentration at the nugget edge, reasonable fatigue strength evaluation of spot welded lap joint is very important to estimate the reliability and durability of th spot welded structure and to establish a criterion of ling life fatigue design. For reasonalbe fatigue strength evaluation, it is necessary to estimate the fatigue strength of spot welded lap joints, systematically. So far, many investigators have numerically and experimentally studied on the systematic fatigue strength estimation for various spot welded lap joints, and the methods suggested has been considerably accumulated. By the way, for applying them in practical fatigue design of the thin sheet structure fabricated by spot welding ,it is also necessary to verify their efficiency and reliability on the predicted results, Therefore, in this study, a statistical fatigue strength estimation method for spot welded lap joints was developed by using the Weibull probability distribution function. From the result, it was found that fatigue strength and fatigue life of the spot welded lap joints having various dimension were able to be statically predicted . And also, a reliable criterion for long life fatigue design of the spot welded lap joint could be established.

• The KIPS Transactions:PartD
• /
• v.9D no.3
• /
• pp.389-398
• /
• 2002

Software reliability growth models are used in testing stages of software development to model the error content and time intervals between software failures. This paper presents a stochastic model for the software failure phenomenon based on a nonhomogeneous Poisson process(NHPP) and performs Bayesian inference using prior information. The failure process is analyzed to develop a suitable mean value function for the NHPP ; expressions are given for several performance measure. Actual software failure data are compared with several model on the constant reflecting the quality of testing. The performance measures and parametric inferences of the suggested models using Rayleigh distribution and Laplace distribution are discussed. The results of the suggested models are applied to real software failure data and compared with Goel model. Tools of parameter point inference and 95% credible intereval was used method of Gibbs sampling. In this paper, model selection using the sum of the squared errors was employed. The numerical example by NTDS data was illustrated.

• Communications for Statistical Applications and Methods
• /
• v.15 no.6
• /
• pp.909-923
• /
• 2008

This paper considers the replacement model and the preventive maintenance model following the expiration of combination warranty for a repairable system. If the system fails after the combination warranty is expired, then it is minimally repaired at each failure. The criterion used to determine the optimal replacement policy and the optimal preventive maintenance policy is the overall value function based on the expected cost rate per unit time and the expected downtime per unit time. The numerical examples are presented for illustrative purpose when the failure time follows a Weibull distribution.

• Journal of Korean Society of Forest Science
• /
• v.90 no.4
• /
• pp.573-581
• /
• 2001

The purposes of this study are to estimate mean willingness to pay (WTP) for preventing forest fires by contingent valuation method (CVM), and to calibrate the variables affecting WTP. The forest fire prevention fund was utilized as a payment vehicle to elicit respondents' willingness to pay (WTP). A total of 500 adults who reside in Seoul Metropolitan area were selected by two-stage cluster sampling and conducted the face-to-face interview. The scenario was designed to meet the requirements for double-bounded dichotomous choice CVM. More than half of the respondents (64.6%) have a willing to pay for the fund. The mean WTP was \4,532. Therefore a total WTP for the population was \34,165,758,000. The calibration of Weibull proportional hazard model showed that education level, environmental conservation intention and negative consciousness about the effect of forest fire were independent variables strongly influencing the WTP.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2009.11a
• /
• pp.154-154
• /
• 2009

본 연구는 고전압 전력기기인 몰드형변압기와 계기용 변성기인 CT, PT 절연특성에 유용한 Epoxy/Nano-Micro Mixture Composites(이하,ENMMC)를 개발하기위해 무엇보다 중요한 것은 Nano입자인 $SiO_2$_10nm입자의 표면을 제어하여 즉, 표면의 소수성을 크게 하여 나노입자의 균질한 분산을 얻은것이 무엇보다 중요하다. 개발된 Epoxy/$SiO_2$_10nm Nanocomposites와 Microcomposites을 기계적 전단응력을 이용하여 균질 혼합을 실시하였다. 이런 조건을 이용한 전기적특성을 측정하기위해 구대구 전극이 완전함침된 평등전계하에서 절연파괴전압을 측정하기 시편을 제조하였다. 마이크로입자의 충진함량을 일정하게 유지하여 나노입자 충진함량비율을 4가지로 변화시켜 절연파괴특성을 연구하였다. 충진함량이 나노입자의 경우 1wt%이하의 값이 상대적으로 우수한 절연파괴특성으로 와이블 플롯을 통하여 알수있었다. 상대적으로 멀티나노콤포지트의 형상파라미터가 큰 결과값을 얻을수 있었다. 그리고 스케일파리미터는 누적확률 밀도함수로서 63.2%에서 대단히 큰 초절연성의 절연소재를 개발할수 있었다.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.15 no.6
• /
• pp.1908-1918
• /
• 1991

Fatigue tests have been carried out to measure the degradation of the residual strength and the fatigue life in carbon/epoxy (0/45/90/-45)$_{2s}$ composite materials. Theoretical predictions of residual strength and fatigue life were compared with experimental results. Distribution characteristics were studied using the probability of failure based on the cumulative distribution function and median rand. The static ultimate strength of carbon/epoxy composites used herein is observed to be relatively higher than that of existing similar composites ; while fatigue life is shorter due to the brittleness of matrix. The fatigue life obtained in these experiments is shorter than that estimated by residual strength degradation model when the stress level above 0.6 For the stress level of 0.6, the experimental value was abruptly increased. The cumulative distribution function for the static ultimate strength is well correlated to that for the strength converted from the measured fatigue life. Also, the predicted distribution of residual strength shows good agreement with the experimental results. Therefore, it is proven that the residual strength degradation model is reasonable.e.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.9
• /
• pp.1325-1330
• /
• 2010

The breakdown of any critical component of a turbine results in the outage of power plants. Unexpected failure decreases equipment utilization and causes enormous economic losses. Currently, we conduct conservative preventive maintenance for a maintenance period that is proposed by a vendor. In the rapidly changing business environment, reliability-based maintenance is required in order to remain competitive and reduce maintenance costs while maintaining the reliability of equipment. In order to determine an appropriate maintenance period for guaranteeing reliability, we must determine the failure probability by carefully analyzing the failure history of the equipment. In this study, we created a database of failure history for power-plant turbines, predicted the best repair time using the Weibull function, and investigated how the appropriate maintenance cycle can be determined.

• Journal of the Korean earth science society
• /
• v.34 no.6
• /
• pp.524-535
• /
• 2013

Catastrophe risk models require the damage functions of each vulnerable item in inventory to estimate volcanic ash losses. The damage functions are used to represent the relation between damage factors and damage and also widely used in engineering and natural hazard studies to calculate the vulnerability. In most cases, damage functions are constructed as fragility or vulnerability curves, and researchers are confused by the similarities between them particularly when they perform interdisciplinary research. Thus, we aim to explain the similarities and differences between fragility and vulnerability curves and their relationship by providing case studies to construct them. In addition, we suggest a simple method to construct the damage functions between damage ratio and volcanic ash thickness using limited damage data. This study comes from the fact that damage functions are generally constructed using damage data. However, there is no available volcanic ash damage data in Korea, and not even enough volcanic disaster data to construct damage functions in the world, compared to other hazards. Using the method suggested in the study and the limited damage data from Japan and New Zealand, we construct Weibull-type functions or linear functions dependent of available data to calculate volcanic ash loss estimation, which we think need to be corrected to make it more suitable for inventory characteristics and environmental conditions in Korea.