• 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 the Korean Data and Information Science Society
• /
• v.20 no.5
• /
• pp.933-939
• /
• 2009

The bivariate data in clinical research fields often has two types of failure times, which are mark variable for the first failure time and the final failure time. This paper showed how to generate bootstrap data to get Bayesian estimation for the joint distribution of bivariate survival times. The observed data was generated by Frank's family and the fake date is simulated with the Gamma prior of survival time. The bootstrap data was obtained by combining the mimic data with the observed data and the simulated fake data from the observed data.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.17 no.32
• /
• pp.221-226
• /
• 1994

Almost preventive maintenance policies assumed that the system after pm has failure rate as before pm with probability p and as good as new with probability 1-p. This paper considers the s-expected cost of the model with imperfect periodic preventive maintenance that increasing minimal repair costs at failure and obtains the optimum periodic preventive maintenance time. Numerical example are shown in which the failure time of the system has gamma distribution.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.1
• /
• pp.1-14
• /
• 2004

This paper surveys some existing direct adaptive feedback control schemes for linear time-invariant systems with actuator failures characterized by the failure pattern that some inputs are stuck at some unknown fixed or varying values at unknown time instants, and applications of those schemes to aircraft flight control system models. Controller structures, plant-model matching conditions, and adaptive laws to update controller parameters are investigated for the following cases for continuous-time systems: state tracking using state feed-back, output tracking using state feedback, and output tracking using output feedback. In addition, a discrete-time output tracking design using output feedback is presented. Robustness of this design with respect to unmodeled dynamics and disturbances is addressed using a modified robust adaptive law.

• Journal of the Korean Geotechnical Society
• /
• v.35 no.12
• /
• pp.45-58
• /
• 2019

In this study, a stochastic analysis procedure based on numerical analysis was proposed to evaluate a kind of intensity-duration rainfall threshold for the initiation of slope failure due to rainfall infiltration. Fragility curves were generated as a function of rainfall intensity-duration from the results of probabilistic slope stability analysis by MCS considering the uncertainty of the soil shear strength, reflecting the results of infiltration analysis of rainfall over time. In the probabilistic analysis, slope stability analyses combined with the infiltration analysis of rainfall were performed to calculate the limit state function. Using the derived fragility curves, a chart showing the relationship between rainfall intensity and slope failure-time was developed. It is based on a probabilistic analysis considering the uncertainty of the soil properties. The proposed probabilistic failure distribution analysis could be beneficial for analyzing the time-dependent failure process of soil slopes due to rainfall infiltration, and for predicting when the slope failure should occur.

• Journal of the Korean Society of Safety
• /
• v.31 no.4
• /
• pp.90-96
• /
• 2016

This study proposes a dynamic reliability analysis of control system as a method of quantitative evaluation of its performance in probabilistic terms. In this dynamic reliability analysis, the failure event is defined as an event that the dynamic response of the structural system exceeds a displacement limit, whereas the conventional reliability analysis method has limitations that do not properly assess the actual time history response of the structure subjected to dynamic loads, such as earthquakes and high winds, by taking the static response into account in the failure event. In this first paper, we discuss the control effect of the viscous damper on the seismic performance of the member-level failure where the failure event of the structural member consists of the union set of time-sequential member failures during the earthquake excitations and the failure probability of the earthquake-excited structural member is computed using system reliability approach to consider the statistical dependence of member failures between the subsequent time points. Numerical results demonstrate that the proposed approach can present a reliable assessment of the control performance of the viscous damper system in comparison with MCS method. The most important advantage of the proposed approach can provide us more accurate estimate of failure probability of the structural control system by using the actual time-history responses obtained by dynamic response analysis.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.7
• /
• pp.428-434
• /
• 2020

The one-shot weapon system is destroyed after only one mission. So, the system requires high reliability. Guided missiles are one-shot weapon systems that have to be analyzed by storage reliability since they spend most of their life in storage. The analysis results depend on the model and the ratio of correct censored data. This study was conducted to propose a method to more accurately predict the future failure rate of Air force guided missiles. In the proposed method, the failure rate is predicted by both MTTF (Mean Time To Failure) and MTBF (Mean Time Between Failure) models and the model with a smaller error from the real failure rate is selected. Next, with the selected model, the ratio of correct censored data is selected to minimize the error between the predicted failure rate and the real failure rate. Based on real field data, the comparative result is determined and the result shows that the proposed sampling rate can predict the future failure rate more accurately.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.35 no.3
• /
• pp.47-62
• /
• 1993

The laboratorv model test was carried out to investigate the behavior of pore water pressure, the critical amount of rainfall for slope failure, the pattern of failure, and the variation of seepage line at the slope with the uniform material of embankment by changing the slope angles and rainfall intensities. The results were was summarised as follows : 1.At the beginning stage of rainfall, the negative pore pressure appeared at the surface of slope and the positive pore pressure at the deep parts. But, the negative one turned into the positive one as the rainfall continued and this rapidly increased about 50 to 100 minutes before the slope failure. 2.The heavier the rainfall intensity, the shorter the time, and the milder the slope, the longer the time took to reach the failure of slope. 3.As the angle of the slope became milder, the critical amount of rainfall for slope failure became greater. 4.Maximum pore water pressure was 10 to 40g/cm$^2$ at the toe of slope and 50 to 90g/cm$^2$at the deep parts. 5.In the respect of the pattern of slope failure, surface failure of slope occurred locally at the toe of slope at the A-soil and failure of slope by surface flow occurred gradually at the top part of slope at the B-soil. 6.As the rainfall continued and the saturation zone in the embankment was formed, the seepage line went rapidly up and also the time to reach the total collapse of slope took longer at the B-soil. 7.As the position of the seepage line went up and the strength parameter accordingly down, the safety factor was 2.108 at the A-soil and 2.150 at the B-soil when the slope occured toe failure. Minimum safety factor was rapidly down to 0.831 at the A-soil and to 0.936 at the B-soil when the slope collapsed totally at the top part of slope.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.21 no.45
• /
• pp.301-307
• /
• 1998

This paper proposes free warranty interval for repairable items when the failure types of item are considered. Failure types are classified into major failure and minor failure. If major failure occurs during warranty period, the item is replaced and if minor failure occurs during warranty period, the item is minimally repaired. This paper determines the optimum free warranty interval which minimizes total expected cost of the free warranty cost model. Numerical example is shown in which failure time of item has weibull distribution.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.4
• /
• pp.250-257
• /
• 2013

This paper introduces a failure analysis procedure that underpins real-time fault prognosis. In the previous study, we developed a systematic eventization procedure which makes it possible to reduce the original data size into a manageable one in the form of event logs and eventually to extract failure patterns efficiently from the reduced data. Failure patterns are then extracted in the form of event sequences by sequence-mining algorithms, (e.g. FP-Tree algorithm). Extracted patterns are stored in a failure pattern library, and eventually, we use the stored failure pattern information to predict potential failures. The two practical case studies (marine diesel engine and SIRIUS-II car engine) provide empirical support for the performance of the proposed failure analysis procedure. This procedure can be easily extended for wide application fields of failure analysis such as vehicle and machine diagnostics. Furthermore, it can be applied to human health monitoring & prognosis, so that human body signals could be efficiently analyzed.