• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.6
• /
• pp.1346-1352
• /
• 2009

Warranty cost of automobile parts varies depending on the parts failure rate in a warranty region of individual markets. Parts failure rate is significantly affected by usage-rate given that other stressors of individual markets are similar. Accordingly, warranty cost can be predicted by failure modeling which reflects usage-rate and using a stochastic process. In this paper, one-dimensional approach is used by applying accelerated failure time model on the assumption that the usage-rate is linear. Such model can explain changes in parts failure rate depending on the changes in usage-rate since it can be expressed as a function of usage-rate. Therefore, acquisition of usage-rate in a new market will automatically lead to estimate of failure rate even without warranty data and warranty cost of parts can be predicted through a renewal process in replacement cases. A case study using warranty data of two real markets is presented in the application part of this paper.

• Proceedings of the Korean Reliability Society Conference
• /
• 2006.05a
• /
• pp.116-125
• /
• 2006

Piston assemblies, which are key components of hydraulic high pressure pumps & motors, are major failure products operating at high pressure and high speed, and the main failure mode is wearout of the shoe surface. To predict the actual life of piston assemblies. we require to find out the most sensitive parameters and establish related empirical formula. In this study, we analyzed the life of piston and shoe assemblies in accordance with variation of speed, pressure, and temperature to reduce the life test time, then analyzed the result of combined accelerated life test which is applied by high speed, speed pressure, and high temperature simultaneously, and finally developed combined accelerated life test model.

• International Journal of Reliability and Applications
• /
• v.18 no.2
• /
• pp.45-63
• /
• 2017

In this paper, we have formulated optimum Accelerated Life Test (ALT) plan for a parallel system with two independent components using masked data with ramp-stress loading scheme and Type-I censoring. Consider a system of two independent and non-identical components connected in parallel. Such a system fails whenever all of its components has failed. The exact component that causes the system to fail is often unknown due to cost and time constraint. For each parallel system at test, we observe its system's failure time and a set of component that includes the component actually causing the system to fail. The stress-life relationship is modelled using inverse power law, and cumulative exposure model is assumed to model the effect of changing stress. The optimal plan consists in finding out the optimum stress rate using D-optimality criterion. The method developed has been explained using a numerical example and sensitivity analysis carried out.

• Journal of Applied Reliability
• /
• v.12 no.3
• /
• pp.187-199
• /
• 2012

In general, accelerated life testing is performed to reduce testing time. But it is difficult to apply accelerated life testing to a system besides components. This paper developed a software which estimates reliability measures of the system from results of accelerated life testing of components building the system. This software can handle the system with a large number of components and complex topology. Multiple failure modes of a component were also considered in this software. Based on the software, reliability measures of a gearbox example at several conditions were estimated from the accelerated life testing results of three components of the gearbox.

• Journal of the Korean Data and Information Science Society
• /
• v.7 no.1
• /
• pp.21-35
• /
• 1996

In this paper, estimation and prediction procedures are discussed for grneral situation in which the failure time follows the independent density $f_{i}({\varepsilon}_{i})$ for the accelerated life testing under Type II censoring. In the context of accelerated life test experiment, procedures are given for estimating the parameters in the Eyring model, and for estimating mean life at a given future stress level. The procedures given are conditional confidence interval procedures, obtained by conditioning on ancillary statistics. A comparison is made of these procedures and procedures based on asymptotic properties of the maximum, likelihood estimates.

• Journal of Korea Port Economic Association
• /
• v.39 no.1
• /
• pp.47-63
• /
• 2023

In smart ports and port automation, the number of vertically deployed container terminals is growing. The purpose of this study is to analyze the productivity of horizontally arranged and vertically arranged container terminals by comparing the main ship operation time, and to recommend future strategies for increasing the operational efficiency of vertically configured container terminals. To achieve our goal, we chose two terminals representating each type, and collected berth allocation status data from 2018 to 2022. Then we analyzed the data using the Accelerated Failure Time (AFT) model, a parametric survival analysis technique. Under the assumption that the working circumstances of the C/C (Container Crane) are the same, we find that the productivity of on-board work of the vertically placed container terminal is higher than that of the horizontally placed container terminal. Our result also shows that the productivity is reduced during the COVID-19 period and the European ships show lower onboard work time. On the basis of these findings, we propose strategies to improve the productivity of vertical container terminals.

• Journal of the Korea Safety Management & Science
• /
• v.17 no.2
• /
• pp.241-247
• /
• 2015

The purpose of the warranty data analysis can be classified into two categories. Two goals is a failure cause analysis and life prediction analysis. In this paper first, we applied multivariate analysis method that can be estimated in consideration of various factors on the failure cause warranty data. In particular, we apply the Tree model and Cox model. The advantage of the Tree is easy to interpret this result as compared to other models. In addition Cox model can quantitatively express the risk. Second, this paper proposed a multivariate life prediction model (AFT) considering a variety of factors. By applying the actual warranty data confirmed the usability.

• Journal of Applied Reliability
• /
• v.18 no.3
• /
• pp.193-200
• /
• 2018

Purpose: In this study, the life of the motor is investigated by performing the accelerated life test with the brush wear of the industrial cleaner motor as the main failure mode. Methods: The accelerating stress factor of the accelerated life test is a voltage, which can increase the number of revolutions of the motor to accelerate the brush wear due to the friction between the brush and the commutator. Also, the accelerating stress level was determined after determining the maximum allowable level of the voltage through the preliminary test. Results: The motor failure time at each accelerating stress level was predicted by regression analysis with brush wear length as performance degradation data. The main failure mode, which is brush wear, of the motor was reproduced by this test. The shape parameter of the Weibull distribution was confirmed to be the same statistically at all accelerating stress levels by the likelihood ratio test. Conclusion: The life of the motor was investigated by performing the accelerated life test with the brush wear of the industrial cleaner motor as the main failure mode. Through the accelerating test method of the cleaner motor, various life expectancy and life expectancy of the acceleration factor are predicted.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.2
• /
• pp.183-194
• /
• 1997

In accelerated life tests, the failure time of an item is observed under a high stress level and based on the time, the failure rates of items we estimated at the normal stress level. In this paper, when the mean of the prior distribution of a parameter is known in Weibull lifetime model with censored failure time data, we study various estimating methods to obtain the empirical Bayes estimator of a parameter from the empirical Bayes approach under the normal stress level by considering the fact that the Bayes estimator is the function of prior parameters and of the acceleration parameter representing the effect of acceleration. And we compare the performance of several empirical Bayes estimators of a parameter in terms of the Bayes risk.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.411-425
• /
• 2021

Accelerated failure time (AFT) model represents a linear relationship between the log-survival time and covariates. We are interested in the inference of covariate's effect affecting the variation of survival times in the AFT model. Thus, we need to model the variance as well as the mean of survival times. We call the resulting model mean and variance AFT (MV-AFT) model. In this paper, we propose a variable selection procedure of regression parameters of mean and variance in MV-AFT model using penalized likelihood function. For the variable selection, we study four penalty functions, i.e. least absolute shrinkage and selection operator (LASSO), adaptive lasso (ALASSO), smoothly clipped absolute deviation (SCAD) and hierarchical likelihood (HL). With this procedure we can select important covariates and estimate the regression parameters at the same time. The performance of the proposed method is evaluated using simulation studies. The proposed method is illustrated with a clinical example dataset.