• Journal of the Korean Data and Information Science Society
• /
• 제16권2호
• /
• pp.401-410
• /
• 2005

In this paper, we calculate the premium rate of reliability insurance policy for brake pads for automobiles using real failure data obtained from use-condition. We try process capability analysis for the manufacturing process of brake-system. We describe the performance factors which have an effect on failure characteristics of brake pads. We also obtain the maximum likelihood estimates of shape and scale parameters of the fitted Weibull distribution for brake pads.

• Journal of the Korean Data and Information Science Society
• /
• 제15권1호
• /
• pp.49-57
• /
• 2004

In this paper, we calculate the premium rate of reliability insurance policy for wiper motors under the assumption of Weibull physics of failure. We also describe the performance factors which have an effect on failure characteristics of wiper motors. The maximum likelihood estimates of shape parameter and scale parameter are obtained by using interval censored real data of sample sizes 6 using MINITAB.

• 콘크리트학회논문집
• /
• 제16권5호
• /
• pp.581-589
• /
• 2004

본 논문에서는 일정진폭 휨인장 및 쪼갬인장 반복하중을 받는 콘크리트의 피로신뢰성과 모델제안을 위하여 도로포장에 사용되는 콘크리트를 대상으로 연구하였다. 실험데이터는 2-모수 Weibull 확률분포함수를 사용하여 신뢰성 해석을 수행하였고, 적합도 검정을 통하여 이를 기반으로 한 모델을 제시하였다. 피로실험은 $150 mm{\times}75 mm$ 시험체의 쪼갬인장피로 실험방법과 $150 mm{\times}150 mm{\times}550 mm$ 시험체의 휨인장피로 실험방법을 적용였으며 일정 진폭 피로하중에 대하여 수행하였다. 두 가지의 실험방법에 대하여 응력 수준 90, 80, $70\%$로 변화하여 실험하고 이때 적용한 응력비는 0.1, 하중재하속도 20Hz, 정현파(sine)를 적용하였다. 연구결과 콘크리트 피로데이터의 분산성으로 인하여 보다 정확한 해석을 위해 두 가지의 해석 기준을 설정하여 제시하였으며, 그래픽방법, 모멘트방법 및 최우도법에 의해 2-모수 Weibull의 매개변수 $\alpha$와 u를 계산하고 이를 이용하여 확률밀도함수(P.D.F)와 누적분포함수(C.D.F)를 도시하였다. 또한, Weibull의 확률분포함수를 이용한 신뢰성 해석은 Kolmogorov-Smirnov의 방법을 사용하여 $5\%$의 유의수준에서 적합도 검정을 수행한 결과 채택되는 것으로 나타났다. 이러한 해석에 기초하여 쪼갬인장피로실험과 휨인장피로 실험에 대한 피로모델을 제시하고 응력수준에 따른 피로수명과 200만회 및 1,000만회에서의 피로강도를 산정하여 제시하였으며 국외의 주요 피로모델과 비교하였다.

• 대한산업공학회지
• /
• 제42권4호
• /
• pp.256-262
• /
• 2016

The Weibull is widely used in reliability analysis, and several studies have attempted to improve estimation of the distribution's parameters. least squares estimation (LSE) or Maximum likelihood estimation (MLE) are often used to estimate distribution parameters. However, it has been proven that Bayesian methods are more suitable for small sample sizes than LSE and MLE. In this work, the Weibull parameter estimation accuracy of LSE, MLE, and Bayesian method are compared for sample sets with 3 to 30 data points. The Bayesian method was most accurate for sample sizes under 25, and the accuracy of the Bayesian method was similar to LSE and MLE as the sample size increased.

• Communications for Statistical Applications and Methods
• /
• 제22권4호
• /
• pp.333-348
• /
• 2015

In this paper, we study a generalization of the modified Weibull distribution. The generalization follows the recent work of Cordeiro et al. (2013) and is based on a class of exponentiated generalized distributions that can be interpreted as a double construction of Lehmann. We introduce a class of exponentiated generalized modified Weibull (EGMW) distribution and provide a list of some well-known distributions embedded within the proposed distribution. We derive some mathematical properties of this class that include ordinary moments, generating function and order statistics. We propose a maximum likelihood method to estimate model parameters and provide simulation results to assess the model performance. Real data is used to illustrate the usefulness of the proposed distribution for modeling reliability data.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제18권2호
• /
• pp.122-129
• /
• 2018

Purpose: This article develops an optimal reliability acceptance sampling plan for the case where the degradation quantity of the performance characteristic follows Weibull distribution. Method: For developing reliability acceptance sampling plans, the sample size and the acceptance constant are determined based on the accelerated characteristic of the test condition and the product. Results: The sample size and the acceptance constant are provided such that the constraints of the producer and the consumer risks are satisfied. Conclusion: Reliability acceptance sampling plans based on the accelerated degradation test method can be used for the quality control within a resonable amount of cost and time. In this article. an optimal reliability sampling plans are newly developed for this purpose.

• 대한전기학회논문지:전력기술부문A
• /
• 제55권3호
• /
• pp.109-115
• /
• 2006

The Weibull distribution is a good candidate for accurate probabilistic model with its rich shape-forming ability and relatively simple CDF(cumulative distribution function). If there are sufficient information to get convincible mean and variance for a probabilistic event, reliable parameters of the Weibull distribution can be determined uniquely. However, sufficient information is not given as usual. There needs more deliberate model building method for that case. This Paper presents an effective parameter estimation technique for Weibull distribution with limited failure data.

• 대한치과기공학회지
• /
• 제43권1호
• /
• pp.13-18
• /
• 2021

Purpose: This study aimed to compare the flexural properties and perform the Weibull analysis of photo-curing three-dimensional (3D) printing resin. Methods: Photo-curing temporary resin (3D polymer) was used as a printing resin. Specimens (65 × 10 × 3.3 ㎣) were prepared following the ISO 20975-1 guidelines and according to the different printing orientations using a digital light processing 3D printer (D2 120; Dentium). The flexural strength (FS), flexural modulus, and work of fracture (WOF) were measured using a universal testing machine (Instron 3344; Instron) at a crosshead speed of 5 mm/min. Results: In this study, the 0° orientation exhibited higher FS and WOF than the 45° orientation. Significant differences were found among the printing orientations (p<0.05). Specimens printed at the 0° orientation were the most accurate. In the Weibull analysis, 0° showed the greatest Weibull modulus (m), which represents a higher reliability. Conclusion: 3D printing should be selected and used by considering flexural properties, size accuracy, and reliability.

• 대한기계학회논문집A
• /
• 제37권9호
• /
• pp.1169-1174
• /
• 2013

대부분의 제품은 다수의 고장모드를 가지지만, 일반적으로 신뢰성 분석에서 고장모드 별로 분석하는 경우는 많지 않다. 또는 신뢰성 분석 시 고장모드에 대한 정보를 제공하지 않거나 대표적인 고장모드로 분석하기도 한다. 특히 신뢰성 인증시험에서 가정하고 있는 형상모수는 제품의 형상모수보다는 고장모드에 대한 형상모수 값이 더 중요하다. 본 연구에서는 다수의 고장모드를 가지는 기계부품에 대한 신뢰성 분석방법으로 경쟁 고장모드 분석방법과 혼합 와이블 분석방법을 소개한다. 그리고 3 가지 고장모드를 가지는 공기압 실린더의 사례에 대해 각 고장모드 별로 형상모수를 구하였으며, 3 가지 고장모드를 고려했을 때와 하나의 고장모드로 가정했을 때의 신뢰성 측도($B_{10}$ 수명, 특성 수명)을 비교하였다.

• 대한전기학회논문지:전력기술부문A
• /
• 제52권7호
• /
• pp.401-406
• /
• 2003

Electric power industries in several countries are currently undergoing major changes, mainly represented by the privatizations of the power plants and distribution systems. Reliable operations of the power plants directly contribute to the revenue increases of the generation companies in such competitive environments. Strategic optimizations should be performed between the levels of the reliabilities to be maintained and the various preventive maintenance costs, which require the accurate estimations of the power plant reliabilities. However, accurate estimations of the power plant reliabilities are often limited by the lack of accurate power plant failure data. A power plant is not supposed to be failed that often. And if it fails, its impact upon the power system stability is quite substantial in most cases, setting aside the significant revenue losses and lowered company images. Reliability assessment is also important for Independent System Operators(ISO) or Market Operators to properly assess the level of needed compensations for the installed capacity based on the availability of the generation plants. In this paper, we present a power plant reliability estimation technique that can be applied when the failure data is insufficient. Median rank and Weibull distribution are used to accommodate such insufficiency. The Median rank is utilized to derive the cumulative failure probability for each ordered failure. The Weibull distribution is used because of its flexibility of accommodating several different distribution types based on the shape parameter values. The proposed method is applied to small size failure data and its application potential is demonstrated.