A ampliação acelerada da demanda por transporte, mais especificamente pelo transporte rodoviário, tem provocado um aumento expressivo no número de acidentes de trânsito nesse ambiente. Consequentemente, a redução dos acidentes de trânsito tem sido um grande desafio para os pesquisadores e gestores da área rodoviária. Porém, os acidentes de trânsito são eventos complexos se considerados os diversos fatores que podem influenciá-los. Dentro desse contexto esta dissertação apresenta um estudo de modelos de previsão de acidentes, que podem ser utilizados para a avaliação do potencial de segurança em determinados locais, identificação e classificação de localidades perigosas ou com propensão a acidentes e avaliação da eficácia de medidas de melhoria da segurança. Nessa dissertação é apresentado um levantamento teórico e metodológico dos modelos de previsão de acidentes, identificando as principais variáveis adotadas bem como as técnicas utilizadas. Para cada modelo revisado foram verificadas as principais diferenças e limitações, e ainda, a análise das variáveis mais influentes presentes nesses modelos. Após, é feita uma comparação de duas abordagens distintas para estimar modelos de previsão de acidentes. A primeira consiste em estimar a ocorrência de acidentes em segmentos da via com as mudanças de características dos elementos de infraestrutura. O segundo relaciona a frequência de acidentes para um único elemento de infraestrutura da via, chamado na literatura internacional de entidade (ex: interseção, curva, tangente, etc.), com base apenas na variável relacionada ao volume de tráfego. O estudo baseado na comparação dessas duas abordagens para a previsão de acidentes revelou que a utilização do volume de tráfego como única variável independente apresenta resultados semelhantes ou até melhores que os modelos baseados em diversos elementos de infraestrutura da rodovia. / The enlargement and the accelerated development of transportation systems, more specifically the land system, have caused the number of road accidents to increase significantly. Therefore, the reduction of road accidents has been a great challenge for researchers and managers in the field of land transportation. However, considering the various factors that may influence them, road accidents are complex events. In this context, this paper presents a study of accident prediction models that can be used to assess the safety potential in certain locations, identify and rank dangerous locations or areas prone to accidents and evaluate the effectiveness of safety improvement measures. Initially, a theoretical and methodological review of accident prediction models is presented, and both the main variables adopted and the methodologies employed are identified. The main differences between all models reviewed and their limitations are presented, and the most influential variables are analyzed. In a second moment, a comparison of two different accident prediction methods is performed. The first method consists in estimating the occurrence of accidents in road sections with changes in the characteristics of infrastructure elements. The second one relates the frequency of accidents based on a single infrastructure element (intersection, curve, tangent, etc.) based on traffic volume only. The study based on the comparison of these two methods found that the use of traffic volume as the only independent variable yields similar or even better results than the models based on various road infrastructure elements.

A ampliação acelerada da demanda por transporte, mais especificamente pelo transporte rodoviário, tem provocado um aumento expressivo no número de acidentes de trânsito nesse ambiente. Consequentemente, a redução dos acidentes de trânsito tem sido um grande desafio para os pesquisadores e gestores da área rodoviária. Porém, os acidentes de trânsito são eventos complexos se considerados os diversos fatores que podem influenciá-los. Dentro desse contexto esta dissertação apresenta um estudo de modelos de previsão de acidentes, que podem ser utilizados para a avaliação do potencial de segurança em determinados locais, identificação e classificação de localidades perigosas ou com propensão a acidentes e avaliação da eficácia de medidas de melhoria da segurança. Nessa dissertação é apresentado um levantamento teórico e metodológico dos modelos de previsão de acidentes, identificando as principais variáveis adotadas bem como as técnicas utilizadas. Para cada modelo revisado foram verificadas as principais diferenças e limitações, e ainda, a análise das variáveis mais influentes presentes nesses modelos. Após, é feita uma comparação de duas abordagens distintas para estimar modelos de previsão de acidentes. A primeira consiste em estimar a ocorrência de acidentes em segmentos da via com as mudanças de características dos elementos de infraestrutura. O segundo relaciona a frequência de acidentes para um único elemento de infraestrutura da via, chamado na literatura internacional de entidade (ex: interseção, curva, tangente, etc.), com base apenas na variável relacionada ao volume de tráfego. O estudo baseado na comparação dessas duas abordagens para a previsão de acidentes revelou que a utilização do volume de tráfego como única variável independente apresenta resultados semelhantes ou até melhores que os modelos baseados em diversos elementos de infraestrutura da rodovia. / The enlargement and the accelerated development of transportation systems, more specifically the land system, have caused the number of road accidents to increase significantly. Therefore, the reduction of road accidents has been a great challenge for researchers and managers in the field of land transportation. However, considering the various factors that may influence them, road accidents are complex events. In this context, this paper presents a study of accident prediction models that can be used to assess the safety potential in certain locations, identify and rank dangerous locations or areas prone to accidents and evaluate the effectiveness of safety improvement measures. Initially, a theoretical and methodological review of accident prediction models is presented, and both the main variables adopted and the methodologies employed are identified. The main differences between all models reviewed and their limitations are presented, and the most influential variables are analyzed. In a second moment, a comparison of two different accident prediction methods is performed. The first method consists in estimating the occurrence of accidents in road sections with changes in the characteristics of infrastructure elements. The second one relates the frequency of accidents based on a single infrastructure element (intersection, curve, tangent, etc.) based on traffic volume only. The study based on the comparison of these two methods found that the use of traffic volume as the only independent variable yields similar or even better results than the models based on various road infrastructure elements.

Esta dissertação apresenta os resultados e análises do monitoramento de um recapeamento asfáltico executado com camada intermediária de geotêxtil. A pista experimental vem sendo solicitada desde 1999, em uma pesquisa que se encontra na sua terceira fase, atingindo 1.103.885 ciclos de carga de 100 kN aplicados. Nessa última fase de estudo o tráfego foi aplicado entre Setembro de 2008 e Fevereiro de 2009, totalizando 121.685 ciclos de carga de eixo de 100 kN. A pista foi dividida em três regiões, sendo empregado em duas delas geotêxteis impregnados com asfalto. Durante o período de estudo foram monitorados parâmetros de resposta estrutural, através de medidas de deflexões e bacias defletométricas, parâmetros de comportamento, através de medidas de deformações permanentes e mapeamento de trincas, e por fim parâmetros ambientais, através das medidas de temperatura do ar e precipitações. Foram também aplicados modelos de previsão de desempenho para o caso de reflexões de trincas. O monitoramento do trincamento na pista mostrou que as regiões com geotêxtil tiveram melhor desempenho quando comparadas à região sem camada intermediária. O comprimento de trincas no final do estudo foi de 764 cm na região sem camada intermediária, 328 cm na região com geotêxtil G150 e 554 cm na região com geotêxtil G150 TF. Foram extraídos corpos de prova da pista, sobre algumas trincas, de modo a se visualizar o modo de desenvolvimento do trincamento, e também para buscar se identificar onde ocorreu trincamento por reflexão. Não foi possível extrair CPs de todas as trincas face a proximidade das mesmas e das dimensões da sonda rotativa, porém do material retirado observou-se ocorrência de trincamento por reflexão, trincamento por fadiga e também trincamento conhecido como top-down cracking (TDC). Os resultados evidenciam que a inclusão de geotêxteis como camada de absorção de tensões aumenta significativamente a vida de recapeamentos asfálticos executados sobre pavimentos trincados. / This dissertation presents the results of the monitoring and analysis of an overlay executed with intermediate geotextile layer. The experimental track has been requested since 1999, in a research that is in its third phase, reaching 1.103.885 cycles of 100 kN axle load applied. In this last phase of the study, the traffic was applied between September 2008 and February 2009, totaling 121.685 cycles of 100 kN axle load. The track is divided into three regions, two of them uses geotextiles impregnated with asphalt. During the study period were monitored parameters of structural response, by means of deflections and basins, parameters of behavior, by means of permanent deformation and mapping of cracks, and finally environmental parameters, through the measures of air temperature and rainfall. Models were also applied to predict the performance of the overlay. The monitoring of cracking showed that regions with geotextile had improved performance when compared to the region without intermediate layer. The length of cracks at the end of the study was 764 cm in the region without intermediate layer, 328 cm in the region with geotextile G150 and 554 cm in the region with geotextile G150 TF. Samples were extracted from the track to see how the cracking was developed and also to identify where there was reflective cracking. On the material removed was observed occurrence of reflective cracking, fatigue cracking and also top-down cracking (TDC). The results show that geotextiles increases the life of concrete asphaltic overlays.

A modelagem de predição de altura comumente exige um amplo conjunto de dados para a etapa de construção e ajuste. Ainda que este tipo de conjunto de dados tenha uma estrutura hierárquica natural, organizada pelas diferentes fazendas, talhões, parcelas, e etc., os modelos de regressão clássicos não consideram a possível variação dos parâmetros, entre os diversos grupos hierárquicos. Os modelos de efeitos mistos, em compensação, podem suportar essa variação, assumindo alguns dos parâmetros dos modelos como sendo estocásticos, além de mostrarem potencial com a possibilidade de diminuição de amostras. Esta técnica permite que a variação interindividual seja explicada considerando parâmetros de efeitos fixos (comuns à população) e parâmetros de efeitos aleatórios (específicos para cada indivíduo). Logo, é natural esperar que em povoamentos florestais com alta variação entre indivíduos, o modelo de efeito misto tenha desempenho superior ao modelo de efeito fixo. Por esta razão, os plantios de Tectona grandis L.f. podem ser considerados como uma população interessante para a modelagem de efeitos aleatórios, uma vez que tal espécie apresenta heterogeneidade de crescimento, sensibilidade à fertilidade e acidez do solo, e a maioria dos seus plantios estabelecidos no Brasil são seminais. Desta maneira este trabalho verifica o ajuste de modelos de efeitos mistos aplicados aos dados de altura total em plantios comerciais de Tectona grandis L.f, localizados no estado do Mato Grosso, com o objetivo na redução do número de amostras quando comparado ao modelo de efeitos fixos. Após a seleção do modelo linear de efeito fixo mais apropriado, testou-se quais dos coeficientes tem efeito aleatório nos diferentes agrupamentos dos dados. Em seguida, selecionou-se o grupo onde o desempenho do modelo de efeito misto em termos de ajuste e predição foi o melhor possível. Por fim, foi verificado a capacidade preditiva dos modelos ajustados por meio de processos de simulação e validação cruzada. Os resultados mostraram que o modelo misto calibrado fornece predições mais confiáveis do que a parte fixa. Este benefício ocorre mesmo ao longo das gradativas diminuições do número de árvores disponíveis para ajuste dentro conjunto de dados teste separados para a calibração do modelo misto. É possível concluir que o modelo calibrado ajustado por talhão, ao invés da parcela, propicia pouca perda de precisão. / Height prediction modeling commonly requires a broad set of data for the construction and adjustment step. Although this type of data set has a natural hierarchical structure, organized by the different farms, plots, plots, etc., the classical regression models do not consider the possible variation of the parameters among the hierarchical groups. The mixed effects models, in compensation, can support this variation, assuming some of the parameters of the models as being stochastic, besides showing potential with the possibility of sample reduction. This technique allows the interindividual variation to be explained considering parameters of fixed effects (common to the population) and parameters of random effects (specific for each individual). Therefore, it is natural to expect that in forest stands with high variation among individuals, the mixed effect model performs better than the fixed effect model. For this reason, the plantations of Tectona grandis L.f. can be considered as an interesting population for the modeling of random effects, since this species presents possible heterogeneity of growth since it is sensitive to the fertility and acidity of the soil, and most of its plantations established in Brazil are seminal. This work verifies the adjustment of mixed effects models applied to total height data in commercial plantations of Tectona grandis L.f, located in the state of Mato Grosso, with the objective of reducing the number of samples when compared to the fixed effects model. After selecting the most appropriate linear model of fixed effect, we tested which of the coefficients have random effect in the different groupings of the data. Then, we selected the group where the performance of the mixed effect model in terms of fit and prediction was the best possible. Finally, the predictive capacity of the adjusted models was verified through simulation and cross-validation processes. The results showed that the calibrated mixed model provides more reliable predictions than the fixed part. This benefit occurs even along the gradual decreases in the number of trees available to fit into separate set of test data for the calibration of the mixed model. It is possible to conclude that the calibrated model adjusted by stand, instead of the plot, provides little loss of precision.

Context. Quality assurance plays a vital role in the software engineering development process. It can be considered as one of the activities, to observe the execution of software project to validate if it behaves as expected or not. Quality assurance activities contribute to the success of software project by reducing the risks of software’s quality. Accurate planning, launching and controlling quality assurance activities on time can help to improve the performance of software projects. However, quality assurance activities also consume time and cost. One of the reasons is that they may not focus on the potential defect-prone area. In some of the latest and more accurate findings, researchers suggested that quality assurance activities should focus on the scope that may have the potential of defect; and defect predictors should be used to support them in order to save time and cost. Many available models recommend that the project’s history information be used as defect indicator to predict the number of defects in the software project. Objectives. In this thesis, new models are defined to predict the number of defects in the classes of single software systems. In addition, the new models are built based on the combination of product metrics as defect predictors. Methods. In the systematic review a number of article sources are used, including IEEE Xplore, ACM Digital Library, and Springer Link, in order to find the existing models related to the topic. In this context, open source projects are used as training sets to extract information about occurred defects and the system evolution. The training data is then used for the definition of the prediction models. Afterwards, the defined models are applied on other systems that provide test data, so information that was not used for the training of the models; to validate the accuracy and correctness of the models Results. Two models are built. One model is built to predict the number of defects of one class. One model is built to predict whether one class contains bug or no bug.. Conclusions. The proposed models are the combination of product metrics as defect predictors that can be used either to predict the number of defects of one class or to predict if one class contains bugs or no bugs. This combination of product metrics as defect predictors can improve the accuracy of defect prediction and quality assurance activities; by giving hints on potential defect prone classes before defect search activities will be performed. Therefore, it can improve the software development and quality assurance in terms of time and cost

Being in the era of Big data, the applicability and importance of data-driven models like artificial neural network (ANN) in the modern statistics have increased substantially. In this dissertation, our main goal is to contribute to the development and the expansion of these ANN models by incorporating Bayesian learning techniques. We have demonstrated the applicability of these Bayesian ANN models in interdisciplinary research including health and cybersecurity. Breast cancer is one of the leading causes of deaths among females. Early and accurate diagnosis is a critical component which decides the survival of the patients. Including the well known ``Gail Model", numerous efforts are being made to quantify the risk of diagnosing malignant breast cancer. However, these models impose some limitations on their use of risk prediction. In this dissertation, we have developed a diagnosis model using ANN to identify the potential breast cancer patients with their demographic factors and the previous mammogram results. While developing the model, we applied the Bayesian regularization techniques (evidence procedure), along with the automatic relevance determination (ARD) prior, to minimize the network over-fitting. The optimal Bayesian network has 81\% overall accuracy in correctly classifying the actual status of breast cancer patients, 59\% sensitivity in accurately detecting the malignancy and 83\% specificity in correctly detecting non-malignancy. The area under the receiver operating characteristic curve (0.7940) shows that this is a moderate classification model. We then present a new Bayesian ANN model for developing a nonlinear Poisson regression model which can be used for count data modeling. Here, we have summarized all the important steps involved in developing the ANN model, including the forward-propagation, backward-propagation and the error gradient calculations of the newly developed network. As a part of this, we have introduced a new activation function into the output layer of the ANN and error minimizing criterion, using count data. Moreover, we have expanded our model to incorporate the Bayesian learning techniques. The performance our model is tested using simulation data. In addition to that, a piecewise constant hazard model is developed by extending the above nonlinear Poisson regression model under the Bayesian setting. This model can be utilized over the other conventional methods for accurate survival time prediction. With this, we were able to significantly improve the prediction accuracies. We captured the uncertainties of our predictions by incorporating the error bars which could not achieve with a linear Poisson model due to the overdispersion in the data. We also have proposed a new hybrid learning technique, and we evaluated the performance of those techniques with a varying number of hidden nodes and data size. Finally, we demonstrate the suitability of Bayesian ANN models for time series forecasting by using an online training algorithm. We have developed a vulnerability forecast model for the Linux operating system by using this approach.

Background: Cesarean delivery rates have been increasing since 1996, and Cesarean delivery is now the most common major operative procedure performed in the United States. Identifying risk factors for wound complications following Cesarean delivery is necessary to prevent unnecessary maternal morbidity. Methods: A case-control study was carried out and data was collected via a medical record review for patients undergoing a Cesarean delivery at the UIHC between 10/1/2011 and 12/31/2012. Results: Several modifiable risk factors were identified, and models based on patient and surgical factors performed better than the current standard NHSN risk index model. Conclusion: More robust prediction models can be created using patient and surgical factors.

The purpose of the study was to find the best prediction models of short-term bank CD ratings using financial variables. This study used short-term bank CD ratings assigned by Moody's and Standard and Poor's.

The diploma thesis focuses on creating a real budiness plan for establishment a porter house, that will offer beer from production of small breweries. Part of the thesis is also a making of prediction models for sales and costs in the branch. It also contains a financial plan and evaluation of the investment effectivity.

The issue with conducting real experiments in design engineering is the cost factor to find an optimal design that fulfills all design requirements and constraints. An alternate method of a real experiment that is performed by engineers is computer-aided design modeling and computer-simulated experiments. These simulations are conducted to understand functional behavior and to predict possible failure modes in design concepts. However, these simulations may take minutes, hours, days to finish. In order to reduce the time consumption and simulations required for design space exploration, surrogate modeling is used. \par Replacing the original system is the motive of surrogate modeling by finding an approximation function of simulations that is quickly computed. The process of surrogate model generation includes sample selection, model generation, and model evaluation. Using surrogate models in design engineering can help reduce design cycle times and cost by enabling rapid analysis of alternative designs.\par Selecting a suitable surrogate modeling method for a given function with specific requirements is possible by comparing different surrogate modeling methods. These methods can be compared using different application problems and evaluation metrics. In this thesis, we are comparing the random forest model and kriging model based on prediction accuracy. The comparison is performed using mathematical test functions. This thesis conducted quantitative experiments to investigate the performance of methods. After experimental analysis, it is found that the kriging models have higher accuracy compared to random forests. Furthermore, the random forest models have less execution time compared to kriging for studied mathematical test problems.