http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
이석훈,강현희 충남대학교 기초과학연구소 1995 忠南科學硏究誌 Vol.22 No.2
We consider a bivariate survival function induced by a random environment for the accelerated life testing model. Gamma and Positive Stable distributions are assumed for the random environment and the poperties of the induced bivariate survival functions especially with respect to the dependence structure due to the environmental effect are investigated. The features of the servival functions according to the parameter values for the dependence and the several stress values are shown in order for the field expert to figure out how the models.
이석훈,김응환 충남대학교 기초과학연구소 2001 忠南科學硏究誌 Vol.28 No.1
We propose a clustering method based on the relationship of characteristic variables of individuals. The main idea of this paper consists of two parts : The one is modeling the appropriate relationship and the membership of each individual. The models in this paper corresponds to the centroids of the conventional clustering algorithms. The major different results from those of the traditional algorithms are to assign physical meaning to the clusters through the interpretation of the models and to provide a classifying rule used for assigning an individual not used in the clustering step to one of the clusters established.
이석훈,박희창 충남대학교 자연과학연구소 1992 忠南科學硏究誌 Vol.19 No.2
We consider a step stress life testing which is deviced for a two-component parallel system with the considerable long life time. The purpose of this paper is to suggest an optimal time for changing the stress when the simple step stress life test is carried out. The life distribution is assumed as the ACBVE by Block and Basu(1974) and the expanded CE model is suggested for the bivariate case. The criterion for optimality is to minimize the variances of the maximum likelihood estimators of the mean life times under the normal stress.
이석훈,윤덕균 한국품질경영학회 2004 품질경영학회지 Vol.32 No.3
In this paper we propose a new method of evaluating customers' service quality using triangle fuzzy numbers and semantic differential scale. These two parameters are useful for quantifying the customers' linguistic variables and subjective judgements. We define the Service Quality(SQ) as the ratio of the expected and perceived service and address the Agreement Rate(AR) calculated from the hamming distance in order to define the Service Rate(SR) as the geometric mean of the SQ and AR. Using the SR in the analysis of service attributes provides more appropriate to measure the customers' service quality and to establish a creative service policy for competitive service quality improvement.