• Journal of Korean Institute of Industrial Engineers
• /
• v.32 no.2
• /
• pp.141-152
• /
• 2006

This paper presents the development of a decision model to examine the optimal repair action for a deteriorating system. In order to make a reasonable decision, it is necessary to perform an analysis of the uncertainties embedded in deterioration and to evaluate the repair actions based on the expected future cost. Focusing on the power law failure model, the uncertainties related to deterioration are analyzed based on the Bayesian approach. In addition, we develop a decision model for the optimal repair action by applying a repair cost function. A case study is given to illustrate a decision-making process by analyzing the loss incurred due to deterioration.

• Communications for Statistical Applications and Methods
• /
• v.7 no.1
• /
• pp.65-86
• /
• 2000

The estimates from an EM when it is applied to a large causal model of 10 or more categorical variables are often subject to the initial values for the estimates. This phenomenon becomes more serious as the model structure becomes more serious as the model structure becomes more complicated involving more variables. In this regard Wu(1983) recommends among others that EMs are implemented several times with different sets of initial values to obtain more appropriate estimates. in this paper a new approach for initial values is proposed. The main idea is that we use initials that are calibrated to data. A simulation result strongly indicates that the calibrated initials give rise to the estimates that are far closer to the true values than the initials that are not calibrated.

• Journal of Korean Society for Quality Management
• /
• v.19 no.2
• /
• pp.63-77
• /
• 1991

Currently the problem of air pollution caused by the motor vehicle emission is one of the most serious problems to be solved. Thus we needed the inspection method and technical innovation constraining the motor vehicle emission. In order to establish the more reasonable certified test, the multivariate sequential rectifying inspection plan designed in this paper has been applied to the domestic vehicles by analyzing the statistic characteristics of the emission distribution. This inspection method is designed to satisfy the evaluation measure constraining domestic vehicle emission, and it serves the defect rectifying system and performance certification of catalytic converts. As the prior parameter for the domestic vehicles, we used the data for the catalytic converts which passed the certified test excuted by the EPK. For the case of engine test, we used those data which passed the certified test of domestic vehicles. The multivariate sequential rectifying inspection plan of the vector parameter is able to minimize the average sample number and increase the pass probability of operating characteristic curve.

• Journal of the Korean Statistical Society
• /
• v.24 no.2
• /
• pp.453-470
• /
• 1995

We describe a constrained optimal classification rule for the case when the prior probability of an observation belonging to one of the two populations is unknown. This is done by suggesting a balanced design for the classification experiment and constructing the optimal rule under the balanced design condition. The rule si characterized by a constrained minimization of total risk of misclassification; the constraint of the rule is constructed by the process of equation between Kullback-Leibler's directed divergence measures obtained from the two population conditional densities. The efficacy of the suggested rule is examined through two-group normal classification. This indicates that, in case little is known about the relative population sizes, dramatic gains in accuracy of classification result can be achieved.

• Communications for Statistical Applications and Methods
• /
• v.20 no.6
• /
• pp.439-454
• /
• 2013

This study presents a Bayesian multiple change-point detection approach to segment and classify the observations that no longer come from an initial population after a certain time. Inferences are based on the multiple change-points in a sequence of random variables where the probability distribution changes. Bayesian multiple change-point estimation is classifies each observation into a segment. We use a truncated Poisson distribution for the number of change-points and conjugate prior for the exponential family distributions. The Bayesian method can lead the unsupervised classification of discrete, continuous variables and multivariate vectors based on latent class models; therefore, the solution for change-points corresponds to the stochastic partitions of observed data. We demonstrate segmentation with real data.

• Computers and Concrete
• /
• v.15 no.2
• /
• pp.279-304
• /
• 2015

In this study, reliability analyses of steel fiber reinforced concrete (SFRC) corbels based on stochastic finite element were performed for the first time in literature. Prior to stochastic finite element analysis, an experimental database of 84 sfrc corbels was gathered from literature. These sfrc corbels were modeled by a special finite element program. Results of experimental studies and finite element analysis were compared and found to be very close to each other. Furthermore experimental crack patterns of corbel were compared with finite element crack patterns and were observed to be quite similar. After verification of the finite element models, stochastic finite element analyses were implemented by a specialized finite element module. As a result of stochastic finite element analysis, appropriate probability distribution functions (PDF's) were proposed. Finally, coefficient of variation, bias and strength reduction (resistance) factors were proposed for sfrc corbels as a consequence of stochastic based reliability analysis.

• Journal of the military operations research society of Korea
• /
• v.9 no.2
• /
• pp.61-68
• /
• 1983

The traditional theories of games are based on an assumption that the payoffs have a single dimension. In reality, any alternative is likely to imply more than one payoff. This paper deals with the game against nature with vector payoffs. The purpose of this paper is to develop methods for finding the practical optimal strategy in the game against nature with vector payoffs. Under the assumption that a prior probability over the stats of nature is given, this paper shows that a practical optimal strategy in this game can be obtained by applying a entropy method in order to assess the payoff weight and by employing the concept of compromise solutions in order to reduce the non-dominated solutions. When subjective payoff weights are unknown as well as known, these methods can be used. A numerical example is given.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.1
• /
• pp.59-74
• /
• 2004

The paper considers the Bayesian interval estimation for the common mean of several normal populations. A Bayesian procedure is proposed based on the idea of matching asymptotically the coverage probabilities of Bayesian credible intervals with their frequentist counterparts. Several frequentist procedures based on pivots and P-values are introduced and compared with Bayesian procedure through simulation study. Both simulation results demonstrate that the Bayesian procedure performs as well or better than any available frequentist procedure even from a frequentist perspective.

• Journal of Korean Institute of Industrial Engineers
• /
• v.43 no.2
• /
• pp.147-153
• /
• 2017

In order to demonstrate a target reliability with a specified confidence level, a new two-stage Bayesian Reliability Demonstration Test (RDT) plans that is known to be more effective than a corresponding single-stage one is proposed and developed by Bayesian framework with beta prior distribution for Weibull life time distribution. A numerical example is provided to illustrate the proposed RDT plans and compared with other non-Bayesian and Bayesian plans. Comparative results show that the proposed Bayesian two-stage plans have some merits in terms of required and expected testing time and probability of acceptance.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.6
• /
• pp.1468-1472
• /
• 1996

In this study, the performance of the CART(Classification and Regression Tree) is compared with that of discriminant analysis method. In most experiments using bank data, discriminant analysis shows better performance in terms of the total cost. In contrast, most experiments using insurance data show that the CART is better than discriminant analysis in terms of the total cost. The contradictory result are analysed by using the characteristics of the data sets. The performances of both the Classification and Regression Tree and discriminant analysis depend on the parameters:failure prior probability, data used, type I error, type II error cost, and validation method.