• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1998.10a
• /
• pp.807-813
• /
• 1998

이 논문에서는 로그정규분포에 대한 베이지안 모형선택방법을 제안한다. 일반적으로 , 모수에 대한 사전정보가 비정보적(noninformative)인 경우, 베이즈 요인(Bayes factor)은 결정할 수 없는 상수를 포함하는 것이 일반적이다. 이 경우, 베이즈 요인을 계산하기 위해 최근 활발히 연구중인 고유 베이즈 요인(Intrinsic Bayes factor)방법을 이용한다. 실제의 자료를 통해 로그정규분포의 적합도 검정에 대한 부분적 베이즈 요인을 계산한다.

• The Korean Journal of Applied Statistics
• /
• v.6 no.1
• /
• pp.173-181
• /
• 1993

Nonparametric empirical Bayes estimation of a distribution function with respect to dirichlet process prior is considered when sample sizes are varying from component to component. Zehnwirth's estimate of $\alpha$(R) is modified to be used in our empirical Bayes problem with non-identical components.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.4
• /
• pp.861-870
• /
• 2005

The Bayesian model selection procedures for the shape parameter of gamma distribution are proposed in order to test that the failure rate of gamma distribution is constant, increasing or decreasing. The encompassing intrinsic Bayes factor by Beger and Pericchi (1996) based on Jeffreys prior for shape parameter is used to investigate the usefulness of the proposed Bayesian model selection procedures via both real data and pseudo data.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.5
• /
• pp.469-475
• /
• 2017

Software reliability is one of the most basic and essential problems in software development. In order to detect the software failure phenomenon, the intensity function, which is the instantaneous failure rate in the non-homogeneous Poisson process, can have the property that it is constant, non-increasing or non-decreasing independently at the failure time. In this study, was compared the reliability performance of the software reliability model using the Landely lifetime distribution with the intensity function decreasing pattern and Erlang lifetime distribution from increasing to decreasing pattern in the software product testing process. In order to identify the software failure phenomenon, the parametric estimation was applied to the maximum likelihood estimation method. Therefore, in this paper, was compared and evaluated software reliability using software failure interval time data. As a result, the reliability of the Landely model is higher than that of the Erlang distribution model. But, in the Erlang distribution model, the higher the shape parameter, the higher the reliability. Through this study, the software design department will be able to help the software design by applying various life distribution and shape parameters, and providing software reliability attributes data and basic knowledge to software reliability model using software failure analysis.

• Journal of the Korea Computer Industry Society
• /
• v.6 no.5
• /
• pp.689-696
• /
• 2005

Finite failure NHPP models presented in the literature exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, Goel-Okumoto and Yamada-Ohba-Osaki model was reviewed, proposes the Kappa(2) reliability model, which can capture the nomotonic decreasing nature of the failure occurrence rate per fault. Algorithm to estimate the parameters used to maximum likelihood estimator and bisection method, model selection based on sum of the squared errors and Kolmogorov distance, for the sake of efficient model, was employed. Analysis of failure using real data set, SYS2(Allen P.Nikora and Michael R.Lyu), for the sake of proposing two parameter of the Kappa distribution, was employed. This analysis of failure data compared with the Kappa model and the existing model using arithmetic and Laplace trend tests, bias tests is presented.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.1 s.39
• /
• pp.45-53
• /
• 2006

Finite failure NHPP models presented in the literature exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, Goel-Okumoto and Yamada-Ohba-Osaki model was reviewed, proposes the $x^2$ reliability model, which can capture the increasing nature of the failure occurrence rate per fault. Algorithm to estimate the parameters used to maximum likelihood estimator and bisection method, model selection based on SSE, AIC statistics and Kolmogorov distance, for the sake of efficient model, was employed. Analysis of failure using real data set, SYS2(Allen P.Nikora and Michael R.Lyu), for the sake of proposing shape parameter of the $x^2$ distribution using the degree of freedom, was employed. This analysis of failure data compared with the $x^2$ model and the existing model using arithmetic and Laplace trend tests, Kolmogorov test is presented.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2003.09a
• /
• pp.249-253
• /
• 2003

강서구를 제외한 부산 전지역에서 1998년도에 조사된 지하수 수질 중 6개 성분(pH, TS, KMnO$_4$, Cl, SO$_4$, NO$_3$-N)에 대한 일반통계분석 결과 pH 성분을 제외하고는 5개 성분의 중앙값이 평균보다 적은 값을 보이는 양성왜도를 보임으로써, 수질오염정도를 분석하기 위해 지시크리깅이라는 비모수적인 지구통계분석기법을 적용하였다. 6개 수질성분에 대해 음용수 기준치를 적용하여 음용가능은 “1”의 값이, 음용불가능은 “0”의 값이 주어졌다. 이렇게 변환된 자료를 이용하여 각 성분별로 실험적인 베리오그램 분석을 실시한 결과 pH, TS, SO$_4$ 성분은 선형모델이 선정되었으며, KMnO$_4$, Cl, NO$_3$-N 성분은 구상형모델이 선정되었다. 본 연구에서는 지시크리깅을 이용하여 6개 성분의 분포도를 작성하고 부산지역의 오염정도를 분석하였다. 지시크리깅기법은 연구지역 전체의 정량적인 분포를 나타내지는 못하지만, 오염의 유.무와 오염의 크기를 정확하게 파악할 수 있으며 또한, 이상치(outlier)가 크게 영향을 미칠 수 있는 통계학적인 오류를 보완할 수 있다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2002.05a
• /
• pp.579-584
• /
• 2002

시스템의 신뢰도는 설계 단계에서부터 중요한 제약 조건이 됨과 동시에 그 사용 단계에서도 지속적인 관찰의 대상이 된다. 특히 원자력 발전소와 길이 안전성이 강조되는 시스템에서 는 한 번의고장으로 치명적인 문제를 야기 시킬 수 있다. 따라서, 신뢰도가 높은 시스템을 구축하기 위한 방안과 함께 시스템의 신뢰도에 대한 수리적인 평가를 보다 합리적인 방법으로 할 수 있는 것에 관한 연구가 필요하다고 할 수 있다. 시스템의 신뢰도 평가는 고장를 분포 함수의 추정에서 출발한다. 시스템의 고장를 분포 함수 추정시, 고장 자료를 이용하여 분포 함수의 모수를 추정하지만 대상 시스템의 고장 자료가 없는 경우 다른 유사 시스템의 고장자료를 이용하여 고장률 분포 함수를 추정하게 된다. 기준의 연구들은 유사 시스템의 고장자료를 이용할 때 베이지안(Bayesian)분석 절차를 이용하였다. 하지만 기존 방법들은 추정 절차에 필요한 우도함수(likelihood function)를 유도시 계산상의 어려움이 많다. 본 연 구에서는 각각의 개발 자료에 대한 우도함수를 유도하여 전체적인 시스템의 우도함수를 유도함으로써 이러한 문제점을 해결할 수 있는 새로운 절차를 제안하였다. 또한 모의실험을 통하여 기존의 방법과의 비교함으로써 본 연구의 타당성을 검증하였다.

• The KIPS Transactions:PartD
• /
• v.9D no.3
• /
• pp.389-398
• /
• 2002

Software reliability growth models are used in testing stages of software development to model the error content and time intervals between software failures. This paper presents a stochastic model for the software failure phenomenon based on a nonhomogeneous Poisson process(NHPP) and performs Bayesian inference using prior information. The failure process is analyzed to develop a suitable mean value function for the NHPP ; expressions are given for several performance measure. Actual software failure data are compared with several model on the constant reflecting the quality of testing. The performance measures and parametric inferences of the suggested models using Rayleigh distribution and Laplace distribution are discussed. The results of the suggested models are applied to real software failure data and compared with Goel model. Tools of parameter point inference and 95% credible intereval was used method of Gibbs sampling. In this paper, model selection using the sum of the squared errors was employed. The numerical example by NTDS data was illustrated.

• Journal of Digital Convergence
• /
• v.11 no.9
• /
• pp.159-166
• /
• 2013

In this paper, make a study decision problem called an optimal release policies after testing a software system in development phase and transfer it to the user. When correcting or modifying the software, finite failure non-homogeneous Poisson process model, presented and propose release policies of the life distribution, half-logistic property model which used to an area of reliability because of various shape and scale parameter. In this paper, discuss optimal software release policies which minimize a total average software cost of development and maintenance under the constraint of satisfying a software reliability requirement. In a numerical example, the parameters estimation using maximum likelihood estimation of failure time data, make out estimating software optimal release time. Software release time is used as prior information, potential security damages should be reduced.