• Journal of the Korean Data and Information Science Society
• /
• 제14권4호
• /
• pp.951-963
• /
• 2003

In this paper, we propose goodness of fit test statistics based on the estimated Kullback-Leibler information functions using the data from three step stress accelerated life test. This acceleration model is assumed to be a tampered random variable model. The power of the proposed test under various alternatives is compared with Kolmogorov-Smirnov statistic, Cramer-von Mises statistic and Anderson-Darling statistic.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2000년도 추계학술발표회 논문집
• /
• pp.235-240
• /
• 2000

In this paper a test of fit for exponentiality and we propose the estimator of Kullback-Leibler Information functions using the data from accelerated life tests. This acceleration model is assumed to be a tampered random variable model. The procedure is applicable when the exponential parameter based on the data from accelerated life tests is or is not specified under null hypothesis. Using Simulations, the power of the proposed test based on use condition of accelerated life test under alternatives is compared with that of other standard tests in the small sample.

• 대한기계학회논문집A
• /
• 제36권7호
• /
• pp.751-757
• /
• 2012

사용조건에서 수명시험을 수행할 경우 시험시간이 많이 소요되므로 사용조건보다 가혹한 스트레스 조건을 추가하여 가속수명시험을 수행하면 시험시간을 단축시킬 수 있다. 수명에 영향을 미치는 스트레스를 선정하여 각 스트레스의 사용조건과 가속조건에서 시험을 수행할 경우 가속 스트레스간에 교호작용이 존재할 수 있다. 그러나 기존의 연구에서는 2가지 이상의 가속 스트레스를 부과하여 가속수명시험을 수행할 경우 교호작용을 고려하지 않은 경우와 가속 수명에 대한 모델식을 제시한 연구들이 대부분이다. 본 논문에서는 기계류 부품의 수명에 영향을 미치는 가속 스트레스를 선정한 후 교호작용을 고려한 시험계획에 따라 가속수명시험을 수행한 자료를 이용하여 가속 계수와 가속 스트레스간의 관계식을 산출하는 방법을 제시하고자 한다. 또한 본 논문에서 제시한 교호작용을 고려한 가속시험방법과 가속 계수를 산출할 수 있는 모델식을 이용하여 복합적인 가속수명시험 수행시 쉽게 적용하고 결과를 산출할 수 있는 기초가 될 것으로 기대된다.

• International Journal of Reliability and Applications
• /
• 제1권2호
• /
• pp.155-165
• /
• 2000

In this paper, we propose some estimators of Kullback- Leibler Information functions using the data from three step stress accelerated life tests. This acceleration model is assumed to be a tampered random variable model. Some asymptotic properties of proposed estimators are proved. Simulations are performed for comparing the small sample properties of the proposed estimators under use condition of accelerated life test.

• 대한산업공학회지
• /
• 제20권1호
• /
• pp.39-51
• /
• 1994

Almost all accelerated life tests assume that no basic failure mechanism changes within the test stresses. But accelerated life test, considering failure mechanism changes, is needed since failure mechanism changes when accelerating beyond the used stress. This paper studies the analysis when the failure mechanism changes within the test stresses. The piecewise linear regression, which the join point of two lines is estimated, is applied In particular, two accelerated life tests, with and without a change in failure mechanism are examined.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 2009년도 춘계학술대회 미래 IT융합기술 및 전략
• /
• pp.107-113
• /
• 2009

This paper deals with noninformative priors for such as Jeffres' prior, reference prior and probability matching prior for scale parameter of exponential distribution when the data are collected in multiple step stress accelerated life tests. We find the noninformative priors for this model and show that the reference prior satisfies first order matching criterion. Using artificial data, we perform Bayesian analysis for proposed priors.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제9권1호
• /
• pp.47-58
• /
• 2009

In this paper, an accelerated life test procedure for a vacuum cleaner motor is proposed. We investigate the failure mechanism of the motor and select some accelerating factors and determine the orifice size as a key accelerating factor. Three stress levels of orifice size are tested and the failure data with censored data are analyzed. The modified accelerating test will reduce the test time in design phase by using the accelerating factor.

• Journal of the Korean Data and Information Science Society
• /
• 제26권1호
• /
• pp.281-288
• /
• 2015

In this paper, the maximum likelihood estimators and confidence intervals for parameters of Weibull distribution are derived under three level constant stress accelerated life tests and the assumption that a log quadratic relationship exits between stress and the scale parameter ${\theta}$. The compound linear plan proposed by Kim (2006) is used to allocate the test units at each stress level, which performed nearly as good as the optimum quadratic plan and had the advantage of simplicity. Some simulation studies are given.

• Journal of the Korean Statistical Society
• /
• 제35권3호
• /
• pp.225-238
• /
• 2006

This paper considers noninformative priors for the scale parameter of exponential distribution when the data are collected in step stress accelerated life tests. We find the Jeffreys' and reference priors for this model and show that the reference prior satisfies first order matching criterion. Also, we show that there exists no second order matching prior in this problem. Some simulation results are given and we perform Bayesian analysis for proposed priors using some data.

• 드라이브 ㆍ 컨트롤
• /
• 제20권3호
• /
• pp.9-14
• /
• 2023

In this paper, the selection of a reliable accelerated life test code for a 2-ton forklift was accomplished by choosing the driving resistance coefficient failure-free test time based on a 10,000-hour B₁₀ life. The overall life and average equivalent load of the vehicle were then calculated based on actual driving test conditions using the selected driving resistance coefficient. The gear train's accelerated life test code was selected by adjusting the equivalent load to a torque and rotation speed that did not exceed 125%(about 75HP) of the vehicle rated power. The safety of the test standards was validated by conducting an actual accelerated life test utilizing the proposed test method in this study and comparing the test result with the corresponding theoretical value. It is anticipated that the reliability of the accelerated life test in this paper will be enhanced, by incorporating actual driving performance data collected directly from the forklift and adjusting the conditions used in developing the accelerated life test code.