• International Journal of Reliability and Applications
• /
• v.10 no.2
• /
• pp.101-107
• /
• 2009

In the accelerated tests, the importance of correct failure analysis must be strongly emphasized. Understanding the failure mechanisms is requisite for designing and conducting successful accelerated life test. Under this presumption, a rational method must be identified to relate the results of accelerated tests quantitatively to the reliability or failure rates in use conditions, using a scientific acceleration transform. Most widely used models for relating the results of accelerated tests quantitatively to the reliability or failure rates in use conditions are an accelerated failure time model and a proportional hazards model. The purpose of this research is to compare the usability of the accelerated failure time model and proportional hazards model in the accelerated life tests.

• Communications for Statistical Applications and Methods
• /
• v.23 no.6
• /
• pp.467-478
• /
• 2016

The accelerated failure time model or accelerated life model relates the logarithm of the failure time linearly to the covariates. The parameters in the model provides a direct interpretation. In this paper, we review some newly developed practically useful estimation and inference methods for the model in the analysis of right censored data.

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.268-275
• /
• 2003

We consider the Bayesian accelerated failure time model. The error distribution is assigned a skewed normal distribution which is including normal distribution. For noninformative priors of regression coefficients, we show the propriety of posterior distribution. A Markov Chain Monte Carlo algorithm(i.e., Gibbs Sampler) is used to obtain a predictive distribution for a future observation and Bayes estimates of regression coefficients.

• Journal of Korean Society for Quality Management
• /
• v.32 no.4
• /
• pp.242-251
• /
• 2004

Several statistical methods are introduced 1=o analyze the accelerated failure time data. Most frequently used method is the log-linear approach with parametric assumption. Since the accelerated failure time experiments are exposed to many environmental restrictions, parametric log-linear relationship might not be working properly to analyze the resulting data. The models proposed by Buckley and James(1979) and Stute(1993) could be useful in the situation where parametric log-linear method could not be applicable. Those methods are introduced in accelerated experimental situation under the thermal acceleration and discussed through an illustrated example.

• Proceedings of the Korean Reliability Society Conference
• /
• 2005.06a
• /
• pp.327-337
• /
• 2005

Accelerated life test models and procedure are developed to assess the reliability of Door switch. The main function of door switch is to operate bulb lamp and fan motor. The accelerated life test method and test equipments are developed using the relationship between stresses and life characteristics of the products. Using the developed accelerated life test method, the parameters of the ALT model and life time distribution are estimated and the reliability of the Door S/W at use condition if assessed. The proposed accelerated life test method and procedure may be extended and applied to testing similar kinds of products to reduce test time and costs of the tests remarkably.

• Journal of the Korean Society of Safety
• /
• v.28 no.6
• /
• pp.11-16
• /
• 2013

This paper presents a study on the analyzing reliability of smoke fire detector using accelerated life test. In general, the smoke fire detector is broken by dust which flow in smoke detection chamber. In order to conduct accelerated life test of smoke fire detector dust is set accelerated factor in this paper. The dust is fly-ash which is test particle 5th regulated by KS A 0090. The dust accelerated level is 60 g, 180 g and 360 g and failure time is measured by smoke sensitivity testing. It is considered to failure of detector if detector don't operate within 30 secconds when subjected to an air stream having a velocity of 20 cm/s~40 cm/s containing smoke with a concentration of 15% of rate of light-response of 1 m. The goodness of fit test and mean life prediction conduct using the failure time. The result show that life distribution fits the weibull distribution for failure time data and the mean lifes calculate 22.5 year in domestic product and 14.7 years in overseas product applied dust stress only.

• Transactions of The Korea Fluid Power Systems Society
• /
• v.2 no.2
• /
• pp.1-7
• /
• 2005

For the reliability evaluation of the track drive unit(TDU), firstly, we analyzed the major failure modes through FMEA(failure mode & effects analysis), FTA(failure tree analysis), and 2-stage QFD(quality function deployment), and then quantitatively determined the priority order of test items. The Minitab analysis was also performed for prediction of life distribution and parameters of TDU by use of field failure data collected from 430 excavators for two years. In addition, we converted the fluctuation load in field conditions into the equivalent load, and for evaluation of the accelerated lift by the cumulative fatigues, the equivalent load is again divided into the fluctuation load by reference of test time. And then, by use of the test method in this paper, the acceleration factor(AF) of needle bearing inside planetary gear which is the most weakly designed part of TDU is achieved as 5.3. This paper presents the quantitative selection method of test items for reliability evaluation, the determination method of the accelerated life test time, and the method of non-failure test time based on a few of samples. And, we proved the propriety of the proposed methods by experiments using a TDU for a 30 ton excavator.

• Journal of Korean Institute of Industrial Engineers
• /
• v.29 no.2
• /
• pp.145-149
• /
• 2003

We consider the estimation of failure rate and acceleration factor under type-I censoring without using acceleration model when testing is conducted in only one highly accelerated condition. Failure times of an item are assumed to be exponentially distributed. It is also assumed that the uncertainty about the acceleration factor, the failure time contraction ratio between accelerated condition and use condition, can be modeled by the uniform or gamma prior distribution of appropriate parameters. We respectively use Bayes and maximum likelihood approaches to estimate acceleration factor and failure rate in the use condition. An example is given to show how the method can be applied.

• Journal of Aerospace System Engineering
• /
• v.10 no.1
• /
• pp.51-58
• /
• 2016

In this paper, we present some results on No-failure accelerated life test of aerial vehicle for reliability demonstration. The design of general accelerated life test consists of the three phases: 1) Estimating normal life test time of a single product from Weibull distribution model; 2) Determining the acceleration factor (AF) by utilizing the relation between the life of mechanical components and the applied torque; 3) Calculating the accelerated life test time, which comes from dividing the estimated normal life test time into AF. Then, we applied the calculated life test time to the real reliability test of the flap actuating system, while considering the requirement specification for mechanical components and operating environment of the actuation system. Real experimental processes and results are presented to validate the theory.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.4
• /
• pp.765-775
• /
• 2010

For the accelerated failure time (AFT) model a lot of effort has been devoted to develop effective estimation methods. AFT model assumes a linear relationship between the logarithm of event time and covariates. In this paper we propose a semiparametric support vector machine to consider situations where the functional form of the effect of one or more covariates is unknown. The proposed estimating equation can be computed by a quadratic programming and a linear equation. We study the effect of several covariates on a censored response variable with an unknown probability distribution. We also provide a generalized approximate cross-validation method for choosing the hyper-parameters which affect the performance of the proposed approach. The proposed method is evaluated through simulations using the artificial example.