• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2003년도 추계학술대회 및 정기총회
• /
• pp.263-266
• /
• 2003

This paper is intended to compare between the Bayesian estimates of hazard rate and the hazard rates of mixed distributions. In estimating hazard rates, especially when the MLE method is used, such difficulties as a lack of data and the existence of censored data make it difficult to estimate the rates. For this reason, the estimates of hazard rate based on the Bayesian approach are introduced. For the simplicity, the exponential and gamma distributions are adopted as a sampling distribution and its natural conjugate prior distribution, respectively.

• 한국신뢰성학회:학술대회논문집
• /
• 한국신뢰성학회 2005년도 학술발표대회 논문집
• /
• pp.121-129
• /
• 2005

In this paper, we consider a periodic preventive maintenance(PM) policy in which each PM reduces the hazard rate of amount proportional to the failure intensity, which increases since the last PM and slows down the wear-out speed to that of new one. And the proportion of reduction in hazard rate decreases with the number of PMs. Our model is similar to $ARI_1$ proposed by Doyen and Gaudoin(2004) in the sense of reduction of hazard rate. Our model has totally different wear-out pattern of hazard rate after PM's, however, and the proportion of reduction depends on the number of PM's. Assuming that the system undergoes only minimal repairs at failures between PM's, the expected cost rate per unit time is obtained. The optimal number N of PM and the optimal period x, which minimize the expected cost rate per unit time are discussed. Explicit solutions for the optimal periodic PM are given for the Weibull distribution case.

• 노동경제논집
• /
• 제26권1호
• /
• pp.111-151
• /
• 2003

본 논문은 한국노동패널을 자료로, 위험도 모형 (hazard model)과 Cox 모형을 이용하여 실엽탈출확률 및 요인분석을 시도한 것이다. 실업탈출확률 분석의 결과, 수도권의 경우 실업률 수준 자체는 높지만 실업기간은 오히려 짧은 것으로 나타났다. 수도권과 비수도권 간의 분절성의 존재가 확인되었으며 경상도가 전라도 및 충청도에 비하여 실업탈출확률이 높게 나타나 경상도가 여타 지방과의 차이가 있다는 점도 확인되었다. 구체적으로는 수도권의 평균 실업기간이 9.29개월인 데 비하여 비수도권은 11.86개월로 수도권에서의 실업기간이 2.5개월(유의수준 0.001) 짧게 나타나 수도권에서의 실업탈출이 훨씬 수월함을 알 수 있었으며, 전라도 경상도 간의 차이를 보면, 평균 실업기간에서 전라도가 10.95개월 경상도가 6.96월로 4개월 정도로 전라도가 길게 유의하게 나타난다. 한편, 실업탈출에 영향을 주는 요인분석 결과를 보면, 대도시 이외의 지역, 비수도권 등 노동시장 취약지역에서는 여성일수록, 근속기간이 길수록, 임금근로자일수록, 제조업, 도소매업에 종사했던 근로자일수록 실업으로부터의 탈출확률이 낮아지고 있다. 위의 사실은 지역별 업종별 노동시장정책이 필요하며 특히 지역별 한계근로자에 대한 특성을 파악하여 새로운 직종이나 산업으로의 이동이 원활히 진행될 수 있도록 노동시장정책이 지역별로 보다 세밀히 수립 집행되어야 함을 시사한다.

• 상하수도학회지
• /
• 제23권3호
• /
• pp.305-313
• /
• 2009

This paper provides a method for evaluating a residual life of water mains using a proportional hazard model(PHM). The survival time of individual pipe is defined as the elapsed time since installation until a break rate of individual pipe exceeds the Threshold Break Rate. A break rate of an individual pipe is estimated by using the General Pipe Break Model(GPBM). In order to use the GPBM effectively, improvement of the GPBM is presented in this paper by utilizing additional break data that is the cumulative number of pipe break of 0 for the time of installation and adjusting a value of weighting factor(WF). The residual lives and hazard ratios of the case study pipes of which the cumulative number of pipe breaks is more than one is estimated by using the estimated survival function. It is found that the average residual lives of the steel and cast iron pipes are about 25.1 and 21 years, respectively. The hazard rate of the cast iron pipes is found to be higher than the steel pipes until 20 years since installation. However, the hazard rate of the cast iron pipes become lower than the hazard rates of the steel pipes after 20 years since installation.

• 한국물환경학회지
• /
• 제23권1호
• /
• pp.87-96
• /
• 2007

In this paper a methodology of identifying individual pipes according to the internal and external characteristics of pipe is developed, and the methodology is applied to a case study water distribution pipe break database. Using the newly defined individual pipes the hazard rates of the cast iron 6 inch pipes are modeled by implementing the proportional hazards modeling approach for consecutive pipe failures. The covariates to be considered in the modeling procedures are selected by considering the general availability of the data and the practical applicability of the modeling results. The individual cast iron 6 inch pipes are categorized into seven ordered survival time groups according to the total number of breaks recorded in a pipe to construct distinct proportional hazard model (PHM) for each survival time group (STG). The modeling results show that all of the PHMs have the hazard rate forms of the Weibull distribution. In addition, the estimated baseline survivor functions show that the survival probabilities of the STGs generally decrease as the number of break increases. It is found that STG I has an increasing hazard rate whereas the other STGs have decreasing hazard rates. Regarding the first failure the hazard ratio of spun-rigid and spun-flex cast iron pipes to pit cast iron pipes is estimated as 1.8 and 6.3, respectively. For the second or more failures the relative effects of pipe material/joint type on failure were not conclusive. The degree of land development affected pipe failure for STGs I, II, and V, and the average hazard ratio was estimated as 1.8. The effects of length on failure decreased as more breaks occur and the population in a GRID affected the hazard rate of the first pipe failure.

• International Journal of Reliability and Applications
• /
• 제2권1호
• /
• pp.49-56
• /
• 2001

In this paper we consider empirical Bayes estimation of the hazard rate and survival probabilities with right censored data under the assumption that the hazard function is constant over the period of observation and the prior distribution is gamma. We provide an estimator of the first derivative of the prior moment generating function that converges at each point to the true value in $L_2$ and use it to obtain, easy to compute, asymptotically optimal estimators under the squared error loss function.

• International Journal of Reliability and Applications
• /
• 제1권2호
• /
• pp.133-143
• /
• 2000

Data from weighted distributions appear, among other situations, when some of the data are missing or are damaged, a case that is important in reliability and life testing. The kernel method for hazard rate estimation is discussed for these data where the basic large sample properties are given. As a by product, the basic properties of the kernel estimate of the distribution function for data from weighted distribution are presented.

• Journal of the Korean Statistical Society
• /
• 제28권3호
• /
• pp.359-370
• /
• 1999

We considered a change-point hazard rate model generalizing constant hazard rate model. This type of model is very popular in the sense that the Weibull and exponential distributions formulating survival time data are the special cases of it. Maximum likelihood estimation and the asymptotic properties such as the consistency and its limiting distribution of the change-point estimator were discussed. A parametric bootstrap method for finding confidence intervals of the unknown change-point was also suggested and the proposed method is explained through a practical example.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 춘계학술대회 논문집 전기기기 및 에너지변환시스템부문
• /
• pp.438-440
• /
• 2003

It is very important to ensure system safety during the process of developing a system. Railway system is also devoting a great portion for the safety. Nowadays many countries leading railway industry have their own system assessment process according to the situation of their train control system. In this Paper, several methods to derive Tolerable Hazard Rate are represented in the railway signalling system the characteristics of those methods are also considered respectively.

• Communications for Statistical Applications and Methods
• /
• 제4권2호
• /
• pp.491-496
• /
• 1997

For the survival data analysis of no covariate the discount survival model is proposed to estimate the time-varying hazard rate and the survival function recursively. In comparison with the covariate case it provide the distributionally explicit evolution of hazard rate between time intervals under the assumption of a conjugate gamma distribution. Also forecasting of the hazard rate in the next time interval is suggested, which leads to the forcecasted survival function.