The estimation of the IRT reliability coefficient and its lower and upper bounds, with comparisons to CTT reliability statistics

김성훈 서울대학교 교육연구소 2010 Asia Pacific Education Review Vol.11 No.2

The primary purpose of this study is to investigate the mathematical characteristics of the test reliability coefficient rho(XX)' as a function of item response theory (IRT) parameters and present the lower and upper bounds of the coefficient. Another purpose is to examine relative performances of the IRT reliability statistics and two classical test theory (CTT) reliability statistics (Cronbach's alpha and Feldt-Gilmer congeneric coefficients) under various testing conditions that result from manipulating large-scale real data. For the first purpose, two alternative ways of exactly quantifying rho(XX)' are compared in terms of computational efficiency and statistical usefulness. In addition, the lower and upper bounds for rho(XX)' are presented in line with the assumptions of essential tau-equivalence and congeneric similarity, respectively. Empirical studies conducted for the second purpose showed across all testing conditions that (1) the IRT reliability coefficient was higher than the CTT reliability statistics; (2) the IRT reliability coefficient was closer to the Feldt-Gilmer coefficient than to the Cronbach's alpha coefficient; and (3) the alpha coefficient was close to the lower bound of IRT reliability. Some advantages of the IRT approach to estimating test-score reliability over the CTT approaches are discussed in the end.

용량이 고려된 통신망의 근사신뢰도 계정에 관한 연구

오영환 광운대학교 신기술연구소 1998 신기술연구소논문집 Vol.27 No.-

본 논문에서는 경로 추적범에 의한 통신망의 근사신뢰도를 계정하기 위한 알고리즘을 제안하였다. 통신망의 단점간 근사신뢰도를 계정하기 위해서 단점간 단순 경로와 최소 컬셀을 결정하고 이에 대해 변수 분리법과 공통 다중 함수식을 이용하였다. 특히 단점간 단순 경로들의 AND 연산에 의해 valid group을 구하고 이에 대해 보수 연산을 하여 근사 신뢰도를 계정하였다. 본 논문에서 제안한 알고리즘은 논리 연산자를 이용하였기 때문에 valid group을 결정하는 문제와 보수 연산을 하는데 있어서 더욱 체계적으로 처리할 수 있는 잇점이 있다. In this paper an algorithm for evaluating the bounded reliability in a communication network is presented by path tracing method, The evaluation Process is consisted of four steps. Evaluation of the bounded terminal reliability. (1) The sets of all simple Paths and minimal cut sets between two nodes in a network are determined. (2) Use the separation variables and common multiple functions. (3) The valid groups between all simple paths are determined by using AND operation. (4) The bounded terminal reliability is evaluated using complement operation between allvalid groups. The presented algorithm revealed that the valid groups are easily determined and complement operations are processed by using logic operator and the process has the merit of systematic operation.

Dominant failure modes identification and structural system reliability analysis for a long-span arch bridge

Xin Gao,Shunlong Li 국제구조공학회 2017 Structural Engineering and Mechanics, An Int'l Jou Vol.63 No.6

Failure of a redundant long-span bridge is often described by innumerable failure modes, which make the structural system reliability analysis become a computationally intractable work. In this paper, an innovative procedure is proposed to efficiently identify the dominant failure modes and quantify the structural reliability for a long-span bridge system. The procedure is programmed by ANSYS and MATLAB. Considering the correlation between failure paths, a new branch and bound operation criteria is applied to the traditional stage critical strength branch and bound algorithm. Computational effort can be saved by ignoring the redundant failure paths as early as possible. The reliability of dominant failure mode is computed by FORM, since the limit state function of failure mode can be expressed by the final stage critical strength. PNET method and FORM for system are suggested to be the suitable calculation method for the bridge system reliability. By applying the procedure to a CFST arch bridge, the proposed method is demonstrated suitable to the system reliability analysis for long-span bridge structure.

Reliability computation technique for ball bearing under the stress-strength model

S. Nayak,B. Seal 한국신뢰성학회 2016 International Journal of Reliability and Applicati Vol.17 No.1

Stress function of ball bearing is function of multiple stochastic factors and this system is so complex that analytical expression for reliability is difficult to obtain. To address this pressing problem, in this article, we have made an attempt to approximate system reliability of this important item based on reliability bounds under the stress strength setup. This article also provides level of error of this item. Numerical analysis has been adopted to show the closeness between the upper and lower bounds of this item.

Reliability bound based on the maximum entropy principle with respect to the first truncated moment

Young Hwa Sung,곽병만 대한기계학회 2010 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.24 No.9

Various reliability methods have been suggested in the literature, but the bound of an estimated reliability has received less attention. The maximum entropy principle is used to obtain the reliability bound with respect to the first moment truncated for the first time. Compared to the previous methods of probability bounding based on given moments, our method is demonstrated to generate a tight upper bound that is practically useful for engineering applications. Numerical examples have shown that a good upper bound of probability of failure is well obtained up to four given moments, but with more moments a divergence problem can occur.

Reliability computation technique for ball bearing under the stress-strength model

Nayak, S.,Seal, B. The Korean Reliability Society 2016 International Journal of Reliability and Applicati Vol.17 No.1

Stress function of ball bearing is function of multiple stochastic factors and this system is so complex that analytical expression for reliability is difficult to obtain. To address this pressing problem, in this article, we have made an attempt to approximate system reliability of this important item based on reliability bounds under the stress strength setup. This article also provides level of error of this item. Numerical analysis has been adopted to show the closeness between the upper and lower bounds of this item.

경사제의 다중 파괴모드에 대한 신뢰성 해석

이철응(Lee, Cheol-Eung) 한국해안해양공학회 2008 한국해안해양공학회 논문집 Vol.20 No.2

다중 파괴모드에 대한 경사제 제체의 시스템적 안정성을 신뢰성 이론으로 해석하였다. 먼저 경사제의 안정성과 관련된 네 개의 단일 파괴모드에 대한 신뢰함수가 수립되었다. 각각의 단일 파괴모드에 대한 AFDA 신뢰성 해석 모형이 개발되었으며, CIAD(1985)의 결과와 비교하여 만족스럽게 검증하였다. 단일 파괴모드에 대한 신뢰성 해석에서는 파괴확률 뿐만 아니라 파괴확률에 영향을 미치는 각 확률변수들의 영향계수를 산정하였다. 한편 경사제의 다중 파괴모드에 대한 신뢰성 해석을 수행하기 위해 세 가지, 일차해석법, 이차해석법 그리고 PNET, 해석법이 적용되었다. 해석 결과에 의하면 일차해석법은 파괴확률을 과다 추정하는 경향이 있다. 따라서 다중 파괴모드를 해석할 때는 각 파괴모드간의 상관성을 고려할 수 있는 이차해석법이나 PNET 해석법을 이용하는 것이 바람직하다. 마지막으로 파괴모드간의 상관성에 기인하는 파괴확률의 추가적인 발생 가능성을 다중 파괴모드 해석을 통하여 확인할 수 있었다. A reliability analysis has been performed to investigate the systematic stability of multi-failure modes of rubble-mound breakwaters. The reliability functions of four different failure modes are established straightforwardly. AFDA(Approximate Full Distribution Approcah) reliability models for each failure modes are directly developed and satisfactorily calibrated through the comparison with CIAD's results. In the reliability analysis of single failure mode, the probabilities of failure are calculated and the influence coefficients of random variables in the failure modes are properly evaluated. Meanwhile, three different models such as uni-modal bounds, bimodal bounds, and PNET are applied to evaluate the probabilities of failure of multi-failure modes for rubble-mound breakwaters. It may be found that uni-modal bounds tend to overestimate the probability of failure of multi-failure modes. Therefore, for the systematic reliability analysis of multi-failure modes, it is recommended to use bi-modal bounds or PNET which consider the correlation between the failure modes for rubble-mound breakwaters. By introducing the reliability analysis of multi-failure modes, it could be possible to find out the additional probabilities of failure occurred by the multi-failure modes of a multi-component system such as rubble-mound breakwaters.

다항상태 부품으로 구성된 다항상태 시스템의 신뢰도 분석에 관한 연구

이종형 ( Chong Hyung Lee ) 한국품질경영학회 2007 품질경영학회지 Vol.35 No.2

Most of systems used in real fields are considered as multistate systems with multistate components. As one of methods for performance evaluation of the system, reliability analysis has been popularly used. In this paper, we propose an improved reliability analysis method which is based on state space decomposition method of Aven (1985). In deriving upper bounds, our method uses sets of unspecified states whereas Aven (1985) excludes sets of unacceptance states. Also, closer lower bounds to an exact reliability are obtained by considering of importance of min-path vectors.

다변량 확률분포함수의 추정을 위한 MKDE-ebd 개발

강영진,노유정,임오강 한국전산구조공학회 2019 한국전산구조공학회논문집 Vol.32 No.1

In engineering problems, many random variables have correlation, and the correlation of input random variables has a great influence on reliability analysis results of the mechanical systems. However, correlated variables are often treated as independent variables or modeled by specific parametric joint distributions due to difficulty in modeling joint distributions. Especially, when there are insufficient correlated data, it becomes more difficult to correctly model the joint distribution. In this study, multivariate kernel density estimation with bounded data is proposed to estimate various types of joint distributions with highly nonlinearity. Since it combines given data with bounded data, which are generated from confidence intervals of uniform distribution parameters for given data, it is less sensitive to data quality and number of data. Thus, it yields conservative statistical modeling and reliability analysis results, and its performance is verified through statistical simulation and engineering examples.

Finite-time resilient reliable sampled-data control for fuzzy systems with randomly occurring uncertainties

Sakthivel, R.,Saravanakumar, T.,Ma, Y.K.,Marshal Anthoni, S. North-Holland 2017 Fuzzy sets and systems Vol.329 No.-

<P>This paper deals with the problem of finite-time passivity-based H-infinity resilient reliable state feedback sampled-data control for Takagi-Sugeno fuzzy systems with randomly occurring uncertainties and gain variations, where the randomly occurring characteristics are designed by stochastic variables satisfying the Bernoulli distribution. The main objective of this paper is to design a resilient reliable sampled-data controller such that the resulting closed-loop system is finite-time bounded and satisfies finite-time mixed H-infinity and passive performance index gamma. Based on the constructed Lyapunov-Krasovskii functional and Wirtinger integral inequality, some delay-dependent sufficient conditions are derived for achieving the required result of the considered systems. Moreover the derived finite-time stability conditions are established in the form of linear matrix inequalities (LMIs) which can be solved by Matlab LMI toolbox. Finally two numerical examples are presented to demonstrate the applicability of the proposed control design technique. (C) 2017 Elsevier B.V. All rights reserved.</P>