Neste trabalho propomos um modelo de regress~ao utilizando a distribuição Weibull modificado, esta distribuição pode ser usada para modelar dados de sobrevivência quando a de função de risco tem forma de U ou banheira. Assumindo dados censurados, é considerado os estimadores de máxima verossimilhança e Jackknife para os parâmetros do modelo proposto. Foram derivadas as matrizes apropriadas para avaliar influiência local sobre os parâmetros estimados considerando diferentes peturbações e também é apresen- tada alguma medidas de influência global. Para diferentes parâmetros fixados, tamanhos de amostra e porcentagem de censuras, varia simulações foram feitas para avaliar a distribuição empírica do resíduo deviance modificado e comparado coma distribuição normal padrão. Esses estudos sugerem que a distribuição empírica do resíduo devianve modificado para o modelo de regressão log-Weibull modificado com dados censurados aproxima-se de uma dis- tribuição normal padrão. Finalmente analisamos um conjunto de dados utilizando o modelo de regressão log-Weibull modificado. Uma nova distribuição de quatro parâmetros é definida para modelar dados de tempo de vida. Algumas propriedades da distribuição é discutida, assim como ilustramos com exemplos a aplicação dessa nova distribuição. Palavras-chaves: Modelo de regressão; Distribuição Weibull modificada; Distribuição weibull modificada generalizada; Análise de sensibilidade; Dados censurados; Análise de resíduo / In this paperwork are proposed a regression model considering the modified Weibull distribution. This distribution can be used to model bathtub-shaped failure rate functions. Assuming censored data, we consider a classic and Jackknife estimator for the parameters of the model. We derive the appropriate matrices for assessing local influence on the parameter estimates under diferent perturbation schemes and we also present some ways to perform global influence. Besides, for diferent parameter settings, sample sizes and censoring percentages, various simulations are performed and the empirical distribution of the deviance modified residual is displayed and compared with the standard normal distribution. These studies suggest that the residual analysis usually performed in normal linear regression models can be straightforwardly extend for a martingale-type residual in log-modifiedWeibull regression models with censored data. Finally, we analyze a real data set under log-modified Weibull regression models. A diagnostic analysis and a model checking based on the deviance modified residual are performed to select an appropriate model. A new four-parameter distribution is introduced. Various properties the new distribution are discussed. Illustrative examples based on real data are also given.

Today, most products are produced by several dependent subprocesses. This paper considers the economic-statistical process control for two dependent subprocesses with two assignable causes following Weibull shock distributions. We construct the individual X control chart to monitor the in-coming quality produced by previous process, and use the cause-selecting control chart to monitor the specific quality produced by current process. By using the charts, we can effectively and economically distinguish which subprocess is out of control. The renewal theorem approach is extended to construct the cost model for our proposed control charts, and the successive quadratic programming algorithm and a finite difference gradient method in the Fortran IMSL subroutine (dnconf) is used to determine the optimal design parameters of the proposed control charts. Finally, we give an example to show how to construct and apply the proposed control charts. Furthermore, the sensitivity analysis illustrates the effects of cost and process parameters on the optimal design parameters and the minimal expected cost per unit time for the proposed control charts.

In this article, we make inferences for the Weibull parameters and propose two test statistics for the comparison of two Weibull distributions based on interval-censored data. However, the distributions of the two statistics are unknown and not easy to obtain, therefore a simulation study is necessary. An urn model in the simulation of interval-censored data was proposed by Lee (1999) to select random intervals. Then we propose a simulation procedure with urn model to obtain approximately the quantiles of the two statistics. We demonstrate an example in AIDS study to illustrate how the tests can be applied to the infection time distributions of AIDS.

Submitted by THAIS CAROLINA SANTOS REIS (thais.carolreis@gmail.com) on 2017-12-23T00:12:53Z No. of bitstreams: 1 dissertacao.pdf: 2133624 bytes, checksum: f25eb167a8e2de286e61755cbfe017ff (MD5) / Approved for entry into archive by Claudia Adriana Spindola null (claudia@fct.unesp.br) on 2018-01-03T15:41:54Z (GMT) No. of bitstreams: 1 reis_tcs_me_prud.pdf: 2133624 bytes, checksum: f25eb167a8e2de286e61755cbfe017ff (MD5) / Made available in DSpace on 2018-01-03T15:41:54Z (GMT). No. of bitstreams: 1 reis_tcs_me_prud.pdf: 2133624 bytes, checksum: f25eb167a8e2de286e61755cbfe017ff (MD5) Previous issue date: 2017-06-23 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Na análise de séries climatológicas, a metodologia conhecida como “análise de frequências” inicia-se, após a verificação da validade de algumas suposições, pela escolha e ajuste de uma distribuição de probabilidade. A etapa mais importante desta análise é a escolha ou seleção da distribuição de probabilidade que melhor descreva o verdadeiro comportamento da variável em estudo. Uma vez adotada uma distribuição de probabilidade que esteja bem ajustada, segundo um ou vários critérios, é de interesse, por exemplo, estimar a probabilidade de que eventos de certa magnitude sejam igualados ou excedidos em T anos. O inverso desta probabilidade é chamado de período de retorno, sendo esta uma medida de extrema importância na avaliação de riscos associados a fenômenos climatológicos. Em princípio, qualquer distribuição de probabilidade com suporte nos números reais positivos pode ser utilizada na descrição do comportamento de séries fluviométricas, pluviométricas, eólicas, entre outras. Em se tratando de séries pluviométricas, formadas, por exemplo, pelas pluviosidades diárias, decendiais, mensais, trimestrais e anuais, as distribuições Gama e Weibull são as mais utilizadas. Nos últimos anos, a partir de métodos específicos, uma infinidade de novas distribuições vêm sendo propostas para a análise de observações contínuas e estritamente positivas, cujas aplicações, em sua grande maioria, restringem-se a dados de sobrevivência e confiabilidade. Nesta dissertação de Mestrado, foram avaliadas as performances das distribuições Odd Weibull, Marshall-Olkin Weibull, Weibull exponenciada e Weibull transmutada como alternativas para as distribuições Gama e Weibull na análise de séries pluviométricas históricas, obtidas de 33 estações meteorológicas do Brasil. Concluiu-se que em um contexto geral a distribuição Weibull foi a mais adequada na descrição da variável “precipitação acumulada mensalmente”, contudo, analisando-se pontualmente cada uma das 33 séries, em nove delas uma das extensões foi classificada como a mais adequada na descrição da variável em questão. Diante disso, tal estudo proporcionou a inserção de algumas das recentes extensões da distribuição Weibull na análise de dados climatológicos. / In the climatological series analysis, a methodology known as “frequency analysis” begins, after the validity of some assumptions, by choice and adjustment of a probability distribution. The most important step of this analysis is the choice or selection of probability distribution that best describes the true behavior of the variable under study. Once a probability distribution, that is well adjusted according to one or several criteria, is adopted, it is of interest, for example, to estimate a probability of events of a certain magnitude that are matched or exceeded in T years. The opposite of this probability is called a return period, which is a measure of extreme importance in the evaluation of risks associated with climatological phenomena. In principle, any probability distribution supported by positive real numbers can be used to describe the behavior of fluviometric, pluviometric and wind series, among others. When it comes to the case of rainfall series, formed, for example, by daily, decendial, monthly, quarterly and annual rainfall, the Gamma and Weibull Distributions are more used. In recent years, from specific methods, a plethora of new distributions are being proposed for an analysis of continuous and strictly positive observations, which applications, for the most part, are restricted to survival and reliability data. In this Master’s dissertation, the performances of the Odd Weibull, Marshall-Olkin Weibull, Exponentiated Weibull and Transmutated Weibull Distributions were evaluated as alternatives for the Gamma and Weibull Distributions in the analysis of historical rainfall series, obtained from 33 meteorological stations in Brazil. It was reasoned that in a general context the Weibull Distribution was published in the description of the variable “monthly cumulative precipitation”, however, analyzing each of the 33 series punctually, in nine of them one of the extensions was classified as the most suitable in the description of the variable in question. Thus, such study provided an insight into some of the recent extensions of the Weibull Distribution in the analysis of climatological data.

Submitted by (ana.araujo@ufrpe.br) on 2016-08-09T14:39:17Z No. of bitstreams: 1 Patricia Silva Nascimento Barros.pdf: 808042 bytes, checksum: 292470d809b5626affec619cf6da1f8a (MD5) / Made available in DSpace on 2016-08-09T14:39:17Z (GMT). No. of bitstreams: 1 Patricia Silva Nascimento Barros.pdf: 808042 bytes, checksum: 292470d809b5626affec619cf6da1f8a (MD5) Previous issue date: 2015-05-29 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / The Weibull distribution is very popular to model survival data. Many modi cations of the Weibull distribution have been proposed in recent years. Inspired by the notion of generalized gamma-generated family of distribution of Zografos e Balakrishnan, the Generalized Weibull Distributions Class is proposed. A comprehensive treatment of the mathematical properties of some new distribution is provided, expressions giving for the distribution function, density function, hazard function, moment generating function, characteristic function, mean, variance, skewness and kurtosis. Real data sets were modeled showing best ts, according the chosen criteria, were obtained by the new models. Thus for the Generalized Weibull Distributions Class, in spite of having more parameters to be estimated, the algorithms were able to best adjust the survival analysis data. / A distribuição Weibull é muito popular para modelar dados de sobrevida. Muitas modificações da distribuição Weibull foram propostas nos últimos anos. Inspirado pela noção da família gama-generalizada de distribuições de Zografos e Balakrishnan, é proposta, a classe de distribuições Weibull generalizada. Um tratamento compreensivo das propriedades matem áticas de algumas novas distribuições é feita, sendo encontradas as expressões para a função de distribuição, função densidade, função de risco, função geradora de momentos, função característica, m édia, variância, assimetria e curtose. Ajustou-se os novos modelos para conjuntos de dados reais veri cando, pelos critérios de escolha, que os melhores ajustes foram obtidos pelos novos modelos. Dessa forma a classe de distribuições Weibull Generalizada mesmo tendo mais parâmetros a serem estimados, os algoritmos foram capazes de ajustar os dados de análise de sobrevivência.

This master thesis is part of a big project at Siemens Industrial Turbomachinery (SIT) in Finspång aimed to use the operation experience available at SIT to predict the state of the gas turbines in general and some mechanical components in particular. The objective of the thesis is to continue the development of a prediction model based on experience data for estimations of a components lifetime. In a previous master thesis by Alessandro Olivi statistical analysis of environmental attributes effect on the expected lifetime of components in a gas turbine was performed. Olivi’s thesis constitutes the starting point on which to keep building to create a reliable prediction model. In this thesis extensive validation tests have been performed in order to further quantify the reliability of the model. Investigations aimed towards finding ways to further develop and improve the prediction model are carried out. The relevant new findings are applied to the model and analysis concerning improvements in the prediction accuracy is carried out. It was revealed that the model is able to make accurate predictions for most of the validation points for each failure mode, but more research is needed to obtain a completely reliable prediction model.

This diploma thesis deals with issue of quality monitoring and final product testing at the AKD line. Output of this thesis is calculation of testing time optimization by Weibull distribution. In the beginning reader is familiar with AKD device, what quality management is and ways how to test final products in electrotechnical manufacturing. Main goal of this thesis is to describe differences between testing methods and optimize Burn-In testing method.

Fiber reinforced polymer composite laminates make up more than 50% of modern aircrafts. Such composite laminates are exposed to various environmental and in-service thermo-mechanical load conditions. Transverse/intralaminar cracking is usually the first form of damage appears in a composite laminate and they tend to increase in number during the service life. The growth in number of these cracks significantly degrades the thermo-elastic properties of the composite laminate and eventually leads to final failure. Thus, it is important to predict the crack density (number of cracks per unit length) growth in both non-interactive crack density region and interactive crack density region and its effect in thermo-elastic properties degradation. Non-interactive crack density region is the region where the cracks are far apart and stress perturbation between cracks do not overlap. Interactive crack density region is where the cracks are close to each other and stress perturbation between cracks overlaps and affects the formation of new cracks. In this study, transverse cracks in thick Glass Fiber Epoxy (GF/EP) cross-ply composite laminates under quasi-static tensile loading and tension-tension fatigue loading have been analyzed and predicted. In the first paper attached here, increase in number of transverse cracks in GF/EP cross-ply laminates under quasi-static tensile loading at room temperature (RT) are analyzed using 2 material systems. The failure stress distribution in 90° plies of the laminates is defined by Weibull distribution and the Weibull parameters are determined from crack density versus applied thermo-mechanical transverse stress in 90° layer (σTCLT) data points within the non-interactive crack density region. The crack density growth is then predicted versus the σTCLT and applied mechanical strain in the laminate from the determined Weibull parameters using Monte Carlo method and the stress distribution models between adjacent cracks. The predicted results using the novel stress distribution model introduced here were in good agreement with the non-interactive and interactive crack density regions of test results. The importance of using the Monte Carlo method and novel stress distribution model to predict the whole crack density region have been emphasized in the article, in addition to that it also redefined the interval of non-interactive crack density region.  The second paper expands the concept from the first paper, to address the tension-tension fatigue loading at RT. It deals with the crack density analysis and prediction in [0/90]s GF/EP laminate under fatigue loading at RT. The fatigue tests were performed at 3 maximum stress levels. Here the Weibull parameters were determined from the data points within the non-interactive crack density region in quasi-static and fatigue loading. From the determined Weibull parameters of each stress level and using Monte Carlo method and the novel stress distribution model, the crack density versus the number of fatigue cycles were predicted and in good agreement with the fatigue test results at the respective stress level. The intention here was to use Weibull parameters of one stress level to predict crack density at arbitrary stress levels. Based on it, the predicted results were not sufficiently good and suggested to revisit the Weibull parameter determination by performing fatigue tests at two stress levels.  In the attached paper 3, new methodology on crack density growth simulation and Weibull parameter determination in tension-tension fatigue loading has been developed. In the newly developed methodology, in detailed fatigue tests are performed at one maximum stress level to obtain all data points and at higher stress level to obtain one data point that is a crack density data point at certain number of cycles to determine Weibull parameters. Using the determined Weibull parameters from non-interactive crack density region, the whole crack density region was successfully predicted for other stress levels.

No description available.

No / The probability density and exceedance probability functions of the induced currents in a screened cable connecting two enclosures, resulting from the close. presence of single and multiple mobile phones working at 900 MHz, are investigated. The analysis of the problem is undertaken using the Method of Moments, but due to weak coupling, the impedance matrix was modified to reduce the memory and time requirements for the problem, to enable it to be executed multiple times. The empirical probability distribution functions (PDFs) and exceedance probabilities for the induced currents are presented. The form of the PDFs is seen to be quite well approximated by a log-normal distribution for a single source and by a Weibull distribution for multiple sources