• Communications for Statistical Applications and Methods
• /
• 제5권3호
• /
• pp.723-731
• /
• 1998

In this paper, we consider a hierarchical Bayes estimation of the parameter, the reliability and hazard rate function based on samples from a Burr type X failure model. Bayes calculations can be implemented by means of the Gibbs sampler and a numerical study us provided.

• Structural Engineering and Mechanics
• /
• 제66권1호
• /
• pp.75-84
• /
• 2018

This study presents the reliability-based analysis of nonlinear structures using the analytical fragility curves excited by random earthquake loads. The stochastic method of ground motion simulation is combined with the random vibration theory to compute structural failure probability. The formulation of structural failure probability using random vibration theory, based on only the frequency information of the excitation, provides an important basis for structural analysis in places where there is a lack of sufficient recorded ground motions. The importance of frequency content of ground motions on probability of structural failure is studied for different levels of the nonlinear behavior of structures. The set of simulated ground motion for this study is based on the results of probabilistic seismic hazard analysis. It is demonstrated that the scenario events identified by the seismic risk differ from those obtained by the disaggregation of seismic hazard. The validity of the presented procedure is evaluated by Monte-Carlo simulation.

• Earthquakes and Structures
• /
• 제19권6호
• /
• pp.471-484
• /
• 2020

Previous earthquakes show that, structural safety evaluations should include the evaluation of nonstructural components. Failure of nonstructural components can affect the operational capacity of critical facilities, such as hospitals and fire stations, which can cause an increase in number of deaths. Additionally, failure of nonstructural components may result in economic, architectural, and historical losses of community. Accelerations and random vibrations must be under the predefined limitations in structures with high technological equipment, data centers in this case. Failure of server equipment and anchored server racks are investigated in this study. A probabilistic study is completed for a low-rise rigid sample structure. The structure is investigated in two versions, (i) conventional fixed-based structure and (ii) with a base isolation system. Seismic hazard assessment is completed for the selected site. Monte Carlo simulations are generated with selected parameters. Uncertainties in both structural parameters and mechanical properties of isolation system are included in simulations. Anchorage failure and vibration failures are investigated. Different methods to generate fragility curves are used. The site-specific annual hazard curve is used to generate risk curves for two different structures. A risk matrix is proposed for the design of data centers. Results show that base isolation systems reduce the failure probability significantly in higher floors. It was also understood that, base isolation systems are highly sensitive to earthquake characteristics rather than variability in structural and mechanical properties, in terms of accelerations. Another outcome is that code-provided anchorage failure limitations are more vulnerable than the random vibration failure limitations of server equipment.

• Journal of Advanced Marine Engineering and Technology
• /
• 제24권2호
• /
• pp.90-96
• /
• 2000

The technology of preventive maintenance and corrective maintenance is widely applied to ships in order to maintain the good voyageable condition. One of the most important fields of marine engineering is to seek the maximum availability and to solve the stochastic maintenance problem such that the cost for corrective maintenance is minimized. Accordingly, for the purpose of making the most suitable maintenance schedule which minimizes the expected cost function, this paper suggests the method to grasp the failure characteristics by the ship's maintenance data that are collected from the past. And, suggests the method to estimate the optimal maintenance interval by using the dynamic programming and the cumulative hazard rate function attained from the maintenance data.

• 디지털산업정보학회논문지
• /
• 제8권2호
• /
• pp.1-10
• /
• 2012

In this study, software products developed in the course of testing, software managers in the process of testing software test and test tools for effective learning effects perspective has been studied using the NHPP software. The finite failure nonhomogeneous Poisson process models presented and the life distribution applied exponential and log shaped type hazard function. Software error detection techniques known in advance, but influencing factors for considering the errors found automatically and learning factors, by prior experience, to find precisely the error factor setting up the testing manager are presented comparing the problem. As a result, the learning factor is greater than autonomous errors-detected factor that is generally efficient model could be confirmed. This paper, a failure data analysis of applying using time between failures and parameter estimation using maximum likelihood estimation method, after the efficiency of the data through trend analysis model selection were efficient using the mean square error and coefficient of determination.

• Nonlinear Functional Analysis and Applications
• /
• 제28권3호
• /
• pp.793-800
• /
• 2023

In this paper, we propose a generalization of Lindley distribution (GLD) via a special structure that is concern with progressively Type-II right censoring and time failure data. We study the modern properties that we have built by such combination, for example, survival function, hazard function, moments, and estimation by non-Bayesian methods. Application on some selected data related to Lindley distribution (LD) and (ED) have been employed to find out the best distribution that can fit data comparing with the GLD.

• 한국방재학회 논문집
• /
• 제4권3호
• /
• pp.25-31
• /
• 2004

철도사면은 예상치 못한 연약지반의 존재로 인해 붕괴될 수 있다. 이 연구는 철도사면의 파괴원인을 규명하고 대책공법 적용을 위한 강도정수를 결정하는데 그 목적이 있다. 사면파괴의 원인을 파악하기 위하여 시추조사, 콘관입시험, 현장베인시험 등이 수행되었다. 또한 연약지반의 강도정수를 구하기 위해 불교란상태로 시료를 채취하여 실내시험을 수행하였다. 현장원위치시험 및 실내시험 결과 사면 파괴의 원인은 과압밀점토의 점진적 파괴에 따른 연약지반 강도저하영역의 전파에 기인한 것으로 밝혀졌다.

• 한국국방경영분석학회지
• /
• 제35권2호
• /
• pp.33-49
• /
• 2009

이 연구는 국방장비의 고장률함수가 욕조곡선을 이루고 있는지를 조사하기 위한 것이다. 국방장비 가운데 K-1 전차를 선택하여 고장확률밀도함수에 대한 고장률함수의 형태를 조사하였다. 또 다른 목적은 전차 운용환경에 따라 고장발생 차이가 발생하는지를 조사하는 것이다. 연구 결과, K-1전차의 고장률은 대체로 와이블분포를 따르고 있음을 보여준다. 와이블분포는 욕조곡선과 같이 시스템의 고장률이 전체 수명주기동안 다른 모양을 가지고 있을 때 유용하다. 전차가 운용을 개시하는 초기시점의 고장률함수는 고장이 많이 발생하다가 시간이 경과함에 따라 점차 낮아지는 와이블분포의 감소형 함수를 따르고 있다. 초기고장 시기를 지나면 고장이 안정적으로 발생하는 임의고장 단계에 이르며 이 시기는 고장률이 일정한 와이블분포를 따르게 된다. 장비의 사용기간이 증가하게 되면 마모로 인한 고장이 점차 증가하며 고장률함수는 증가형 함수를 따르게 된다. 또한 연구를 통해 전차의 운용 지형이 열악한 환경에서 고장이 더 많이 발생하고 있음을 알 수 있다. 연구 결과 시사점으로는 창정비 주기를 설정하는데 있어 욕조곡선에서 마모고장이 발생하는 시점을 창정비 시기로 선택해야 할 것이다. 또한 전차 운용 지형을 고려하여 창정비 주기를 조절해야 할 것이다.

• Communications for Statistical Applications and Methods
• /
• 제17권1호
• /
• pp.99-106
• /
• 2010

In this paper, we present two methods for obtaining prediction intervals for the times to failure of units censored in multiple stages in a progressively censored sample from proportional hazard rate models. A numerical example and a Monte Carlo simulation study are presented to illustrate the prediction methods.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2007년도 춘계학술대회 논문집
• /
• pp.936-940
• /
• 2007

This paper is about the study on the verification method for railway system SIL which is frequency of hazard, composing Risk, one of the measurement standards for railway system safety. Frequency of hazard can be identified by using FMECA, or HAZOP, and the assessment of identified dangerous failure rate should be done by systematic methods such as FTA. Therefore, this paper provides the hazard identification level for SIL verification and the requirements necessary to verify the integrity of analysis activity.