No description available.

General Purpose computing on Graphic Processing Unit (GPGPU) gained atten- tion in 2006 with NVIDIA’s first Tesla Graphic Processing Unit (GPU) which could perform high performance computing. Ever since, researchers have been working on software and hardware techniques to improve the efficiency of running general purpose applications on GPUs. The efficiency can be evaluated using metrics such as energy consumption and throughput and is defined based on the requirements of the system. I define it as obtaining high throughput by consuming minimum energy. GPUs are equipped with a large number of processing units, a high memory bandwidth, and different types of on-chip memory and caches. To run efficiently, an application should maximize the utilization of GPU resources. Therefore, a good correspondence between the computing and memory resources of the GPU and those of application is critical. Since an application’s requirements are fixed, the GPU’s configuration should be tuned to these requirements. Having models to study and predict the power consumption and throughput of running a GPGPU application on a given GPU configuration can help achieve high efficiency. The main purpose of this dissertation is to find a GPU configuration that best matches the requirements of a given application. I propose three models that predict a GPU configuration that runs an application with maximum throughput while consuming minimum energy. The first model is a fast, low-cost and effective approach to optimize resource allocation in future GPUs. The model finds the optimal GPU configuration for different available chip real-estate budgets . The second model considers the power consumption and throughput of a GPGPU application as functions of the GPU configuration parameters. The proposed model accurately predicts the power consumption and throughput of the modeled GPGPU application. I then propose to accelerate the process of building the model using optimization techniques and quantum annealing. I use the proposed model to explore the GPU configuration space of different applications. I apply multiobjective optimization technique to find the configurations that offer minimum power consumption and maximum throughput. Finally, using clustering and classification techniques, I develop models to re- late the power consumption and throughput of GPGPU applications to the code attributes. Both models could accurately predict the optimum configuration for any given GPGPU application. To build these models I have used different machine learning techniques and optimization methods such as Pareto Front and Knapsack optimization problem. I validated the model produced results with simulation results and showed that the models make accurate predictions. These models could be used by GPGPU programmers to identify the architectural parameters that most affect an application’s power consumption and throughput. This information could be translated into software optimization opportunities. Also, these models can be implemented as part of a compiler to help it to make the best optimization decisions. Moreover, GPU manufacturers could gain insight on architectural parameters which would profit GPGPU applications the most in terms of power and performance and hence invest on these. / Graduate

É crescente a demanda por produtos do mercado verde em todos os setores da Indústria. Na Construção Civil, no entanto, não são conhecidos os atributos do produto edificação sustentável que sejam valorizados pelos clientes. O objetivo desse trabalho, portanto, é identificar os atributos de desenvolvimento sustentável para empreendimentos habitacionais valorizados por clientes finais. O método de pesquisa empregado consistiu em um levantamento (survey) aplicado aos moradores de empreendimentos habitacionais de duas empresas construtoras/incorporadoras da grande Porto Alegre. Para tanto, a pesquisa divide-se em três etapas: preparatória, coleta de dados e análise de dados. A primeira fase da pesquisa corresponde à preparação da ferramenta para questionário, através das análises de sistemas de certificação ambiental para habitações; entrevistas de avaliação pós-ocupação de habitações mais sustentáveis; workshop em sala de aula de grupo de estudantes de edificações sustentáveis e aplicação de dois estudos pilotos. Na etapa dois, coleta de dados, o objetivo do questionário aplicado em clientes finais de edificações habitacionais genéricas - as quais poderiam ou não conter atributos de desenvolvimento sustentável - é a captação de suas opiniões sobre os atributos da lista resultante da etapa preparatória. A meta é entender se esses atributos seriam encarados positiva ou negativamente quando houver a defrontação das pessoas com uma nova necessidade ou possibilidade de compra de imóvel. Além do levantamento, foram realizadas entrevistas com profissionais das construtoras a fim de verificar o conhecimento possuído por eles sobre a opinião de seus clientes. Na etapa de resultados, a análise dos dados foi composta por duas fases: análise quantitativa dos dados obtidos nos questionários aplicados com clientes finais e análise qualitativa, em que é realizado o cruzamento dos resultados da análise quantitativa com os dados das entrevistas realizadas com os profissionais. Dentro da análise quantitativa, são duas as subfases: análise descritiva geral dos dados, cujo objetivo foi identificar as diferentes opiniões de forma conjunta e análise estatística, através da análise fatorial e análise de agrupamentos, cujos objetivos são explorar a relação entre variáveis e agrupar os respondentes conforme suas opiniões. Com a análise de dados, observa-se, de uma forma geral, opiniões positivas em relação aos atributos. Entretanto, percebe-se uma barreira na aceitação de alguns deles, consequência da falta de disseminação de alguns conceitos. / The demand for green products is increasing in all sectors of the industry. Inspite of that, in the construction sector, the attributes of a sustainable building valued by the final clients are not yet known. The goal of this research is to identify attributes of sustainable development for housing projects which are valued by final clients. The research method is a survey with residents of two housing projects built by two construction firms in the Porto Alegre metropolitan area. The research is divided in three phases: preparatory, data collection and data analysis. The first phase is related to the conception of the data collection tool through the analysis of environmental certifying systems for housing, interview for post occupancy evaluation of more sustainable houses and a workshop in a sustainable construction course and development of two pilot studies. In the data collection phase, the objective of the questionnaire applied to final clients of a generic housing (which could or could not have attributes of sustainable development) was to capture their opinion regarding the list of attributes generated in the first phase of the research. The goal was to understand if people when considering the need or possibility of acquiring a new house would view these attributes either as positive or negative. Besides the survey, interviews with construction firms were performed to verify the knowledge of those in the subject of their client's opinion. The data analysis phase was composed by two phases: quantitative analysis of the data obtained in the survey and qualitative analysis of the data obtained in the interviews with professionals. Afterwards these two data types are cross-analyzed. In the quantitative analysis there were two phases: a general descriptive analysis of the data (aiming at identifying different opinions) and the statistical analysis through factor analysis and cluster analysis, whose goal was to explore the relations between the variables and grouping of the respondents, according to their opinion. In general, positives opinions of the attributes were observed. Nonetheless, some constraints in the acceptance of some attributes were also perceived, due to a lack of dissemination of some concepts.

Esta tese apresenta a modelagem e o monitoramento de características de qualidade do tipo fração, que apresentam valores restritos ao intervalo [0,1] A motivação inicial é que a característica de qualidade do tipo fração pode ser melhor aproximada pela distribuição de probabilidade Beta e, geralmente, a modelagem e o monitoramento utilizados não supõem tal distribuição. Desta forma, os objetivos desta tese são: (i) Propor uma sistemática para seleção de modelos paramétricos considerando a classificação das variáveis resposta e das variáveis explicativas; (ii) Propor uma modelagem da fração não conforme com enfoque no modelo de regressão Beta e no modelo linear generalizado; (iii) Propor uma nova carta de controle (CC), Carta Beta, para monitorar características de qualidade do tipo fração (p), com adaptação dos limites de controle a distribuição Beta; (iv) Propor uma CC baseada em modelos Beta para monitorar características de qualidade do tipo fração em função das variáveis de controle do processo; e (v) Propor uma CC baseada em modelos Beta por componentes principais para monitorar características de qualidade do tipo fração em função das variáveis de controle multicolineares. As cartas de controle propostas foram avaliadas através da aplicação de estudos de caso, exemplos retirados da literatura e comparação do número médio de amostras (NMA) via simulação de Monte Carlo. Os modelos de regressão propostos para modelagem de características de qualidade do tipo fração apresentam flexibilidade de escolha da função de ligação, bom desempenho no ajuste e precisão das estimativas dos efeitos das variáveis. As cartas de controle propostas detectaram mais rapidamente todas as alterações induzidas, apresentando melhor desempenho em comparação as outras cartas similares encontradas na literatura. Conclui-se que as ferramentas propostas são adequadas para a modelagem e o monitoramento de características de qualidade do tipo fração. / This thesis presents the modeling and monitoring of the type fraction quality characteristics, which have values confined in the [0,1]-interval. The main motivation is that the type fraction quality characteristics can be better approximated by the Beta probability distribution and generally used for modeling and monitoring not assumes this distribution. Thus, the objectives are: (i) propose a systematic for selecting parametric models considering the classification of the response and explanatory variables; (ii) propose a modeling of the fraction nonconforming items using the Beta regression model and generalized linear model; (iii) propose a new control chart (CC), Beta Charts, to monitor the type fraction quality characteristics (p), with adaptation of the control limits based in the Beta distribution; (iv) propose a Beta model-based to monitor the type fraction quality characteristics adjusted by control variables of the process, and (v) propose a Beta model-based control charts by principal components to monitor the type fraction quality characteristics adjusted by control variables multicollinearity. The proposed control charts were evaluated by application of case studies, numeric examples taken from literature and comparison the average run length (ARL) using Monte Carlo simulation. The proposed regression models for modeling of the type fraction quality characteristics have shown flexibility to choose link function, good performance on the fit and accuracy of estimates of the effects. The proposed control charts detected more quickly all changes induced, showing better performance compared to other similar control charts defined in literature. It is concludes that the proposed tools are appropriate for modeling and monitoring of the type fraction quality characteristics.

Submitted by Guilherme Balieiro (gui_balieiro@yahoo.com.br) on 2018-09-18T01:02:10Z No. of bitstreams: 1 Dissertação.GuilhermeBalieiro.pdf: 631486 bytes, checksum: 18d1e6ca9bef6935b7003b035c365726 (MD5) / Approved for entry into archive by Neli Silvia Pereira null (nelisps@fcav.unesp.br) on 2018-09-18T13:20:37Z (GMT) No. of bitstreams: 1 balieiro_g_me_jabo.pdf: 631486 bytes, checksum: 18d1e6ca9bef6935b7003b035c365726 (MD5) / Made available in DSpace on 2018-09-18T13:20:37Z (GMT). No. of bitstreams: 1 balieiro_g_me_jabo.pdf: 631486 bytes, checksum: 18d1e6ca9bef6935b7003b035c365726 (MD5) Previous issue date: 2018-08-24 / Objetivo: identificar e analisar os atributos da capacidade empreendedora que geram o aproveitamento de oportunidades de negócios através da ação, segundo conceitos da Teoria do Empreendedorismo, presentes no indivíduo empreendedor no setor de cafés especiais. Procedimentos de Pesquisa: realizou-se pesquisa de campo, de caráter qualitativa, com foco no indivíduo empreendedor do setor de cafés especiais. Por meio da aplicação de questionários semiestruturados, coletou-se dados destes empreendedores, que foram submetidos, posteriormente, à análise de conteúdo. Além disso, utilizou-se do Global Entrepreneurship Monitor (GEM) para evidenciar as características de empreendimentos no Brasil e confrontá-las com os dados obtidos no estudo deste setor. Resultados e Discussões: os resultados demonstraram que os empreendedores do setor de cafés especiais utilizam de seus atributos da capacidade empreendedora para a identificação e aproveitamento das oportunidades no mercado. Isso ocorre devido à ação empreendedora, com a construção de conexões e adaptações em ambientes de incerteza. Além disso, em relação ao GEM, algumas diferenças quanto ao gênero e nível de escolaridade foram evidenciadas na composição de perfil do empreendedorismo brasileiro. Conclusões e Limitações da Pesquisa: todos os atributos levantados na literatura para esta pesquisa estiveram presentes nos nove empreendedores analisados, impactando diretamente sua maneira de agir perante o mercado em que estão inseridos. Cabe aqui mencionar que foi analisada uma pequena parcela de empreendedores presentes no mercado de cafés especiais, situados em diferentes regiões do país. Tal fato limita a generalização dos resultados alcançados, pois diferenças podem existir dependendo das práticas utilizadas por cada empreendedor de diferentes perfis e regiões do Brasil. Originalidade: o agronegócio brasileiro é importante segmento econômico, o qual é influenciado diretamente pelos empreendedores frente a suas empresas. Assim, os atributos empreendedores, ofertados por uma Teoria do Empreendedorismo, criam diferenciais competitivos para iniciar e estabelecer uma empresa, principalmente em mercados de nicho, como aqui destacado o mercado de Cafés Especiais. / Purpose: to identify and analyze the attributes of the entrepreneurial capacity that generate the use of business opportunities through the action, according to concepts of Entrepreneurship Theory, present in the individual entrepreneur in the sector of specialty coffees. Design/Approach: a qualitative field research was developed, focused on the individual entrepreneur of the specialty coffee sector. Through the application of semi-structured questionnaires, data were collected from these entrepreneurs, who were subsequently submitted to content analysis. In addition, the Global Entrepreneurship Monitor (GEM) was used to highlight the characteristics of entrepreneurship in Brazil and to compare them with the data obtained in the study of this sector. Findings: the results showed that entrepreneurs in the specialty coffee sector use their attributes of entrepreneurial capacity to identify and take advantage of opportunities in the market. This occurs through the entrepreneurial action, with the construction of connections and adaptations in environments of uncertainty. It was concluded that there are some differences regarding the gender and level of schooling, among the profile of Brazilian entrepreneurship, according to GEM, with a sample of this research. Management Implication: all the attributes raised in the literature for this research were present in the nine entrepreneurs analyzed, directly impacting their way of acting before the market in which they are inserted. It should be mentioned that a small number of entrepreneurs present in the specialty coffee market, located in different regions of the country, were analyzed. This fact limits the generalization of the results achieved, because differences may exist depending on the practices used by each entrepreneur from different profiles and regions of Brazil. Originality: the Brazilian agribusiness is an important economic segment, which is directly influenced by the entrepreneurs in front of their companies. Thus, entrepreneurial attributes, offered by an Entrepreneurship Theory, create competitive differentials to start and establish a company, especially in niche markets, such as the Special Coffee market.

Master of Science / Department of Geology / Abdelmoneam Raef / Seismic reservoir characterization and prospect evaluation based 3D seismic attributes analysis in Kansas has been successful in contributing to the tasks of building static and dynamic reservoir models and in identifying commercial hydrocarbon prospects. In some areas, reservoir heterogeneities introduce challenges, resulting in some wells with poor economics. Analysis of seismic attributes gives insight into hydrocarbon presence, fluid movement (in time lapse mode), porosity, and other factors used in evaluating reservoir potential. This study evaluates a producing lease using seismic attributes analysis of an area covered by a 2010 3D seismic survey in the Morrison Northeast field and Morrison field of Clark County, KS. The target horizon is the Viola Limestone, which continues to produce from seven of twelve wells completed within the survey area. In order to understand reservoir heterogeneities, hydrocarbon entrapment settings and the implications for future development plans, a seismic attributes extraction and analysis, guided with geophysical well-logs, was conducted with emphasis on instantaneous attributes and amplitude anomalies. Investigations into tuning effects were conducted in light of amplitude anomalies to gain insight into what seismic results led to the completion of the twelve wells in the area drilled based on the seismic survey results. Further analysis was conducted to determine if the unsuccessful wells completed could have been avoided. Finally the study attempts to present a set of 3D seismic attributes associated with the successful wells, which will assist in placing new wells in other locations within the two fields, as well as promote a consistent understanding of entrapment controls in this field.

The predominating theories on non-product-related attributes stretch only so far to explain why some brands do not manage to succeed on foreign markets where the product needs are essentially similar to their domestic markets. Therefore, the aim of this paper is to cast new light on the theoretical field of non-product-related attributes, by unveiling which non-product- related attributes affect consumer brand preferences on similar markets. This is done by studying consumer expectations on cosmetics brands in general and brand attitudes towards a Finnish cosmetics brand, on the Swedish market. The results suggest that most of the non-product-related attributes that are covered by predominating theories are still relevant for the formation of consumer preferences, but that new important attributes have also emerged. It was also found that the level of importance varies from attribute to attribute, and that some factors affect consumer preferences more than others. The findings serve as a basis for re-evaluating and expanding the theory of non-product- related attributes, and can help brands to succeed on similar markets.

Currently, auditors face changes in the auditing industry, which has increased the focus on marketing activities. Despite professional ideals, an increase in the commercialization of audit firms has been observed, shaping the identity of the auditor. Auditor identity has been found to be a driver for commercialization, but certain studies find commercialization to be negative for audit practice. We argue that individual auditor attributes can explain orientations in the auditor identity, which in turn could have implications for audit practice. The purpose of this study is to describe and analyze how auditor attributes are related to professional and commercial orientations, and how the various orientations are related to audit practice. The theories used in this study are agency theory, social identity theory, and the theory of professions. We employ a quantitative survey method, with a questionnaire sent out by e-mail including questions about attributes, professional and commercial orientation, and reduced audit quality acts. Our results indicate that a positive relation to professional orientation exists for attributes adaptability and marketing skills. A positive relation for commercial orientation exists for the attribute marketing skills, while a negative relation exists for communication skills. Our results suggest that commercial orientation has a positive relation to auditors’ propensity to engage in RAQ acts. We find that knowledge has a negative relation to RAQ acts. A weak positive relation exists between adaptability and RAQ acts. The degree to which these results explain professional and commercial orientation and RAQ acts is small, why more research is needed.

Made available in DSpace on 2014-06-11T19:23:11Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-02-10Bitstream added on 2014-06-13T20:50:13Z : No. of bitstreams: 1 reis_fc_me_jabo.pdf: 1415727 bytes, checksum: a4392af51a99fc76cb5b4fdf7237317d (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Fundação para o Desenvolvimento da UNESP (FUNDUNESP) / O trabalho foi desenvolvido na Microbacia Hidrográfica do Córrego da Fazenda Glória no Município de Taquaritinga (SP), com o objetivo de avaliar a variabilidade espacial dos atributos físico-químicos de um Argissolo Vermelho-Amarelo em área de cana-de-açúcar, pastagem e vegetação nativa. A amostragem do solo foi feita nos pontos de cruzamento de uma malha regular de 10 m, totalizando 110 pontos na área de cana-de-açúcar, 126 pontos na área de pastagem e 88 pontos na área de vegetação nativa. Os resultados das análises dos atributos do solo foram submetidos à estatística descritiva dos dados, a avaliação de dependência espacial por meio de técnicas geoestatísticas e a interpolação dos dados por meio da krigagem para geração de mapas. Os resultados mostraram que: a) os atributos do solo estudados apresentaram uma estrutura de dependência espacial moderada e forte, com exceção do Ca, Mg, SB e V na área de cana-de-açúcar, da macroporosidade na área de pastagem e da microporosidade, porosidade total, Mg, SB e CTC na área de vegetação nativa; b) os atributos físico-químicos ajustaram-se aos modelos esférico e exponencial, e alguns apresentaram semivariograma sem estrutura definida; c) a variabilidade espacial dos atributos físico-químicos forneceu a visualização de zonas homogêneas de manejo, permitindo a adoção de um sistema de agricultura de precisão; d) vegetação nativa contribui para a melhoria dos atributos físicos e da matéria orgânica em relação às áreas de cana-de-açúcar e pastagem; e) as maiores variabilidades medidas por meio do coeficiente de variação foram observados na área da vegetação nativa; f) as técnicas geoestatísticas forneceram um maior detalhamento das áreas em estudo, possibilitando em futuras amostragens usar o alcance de dependência espacial para o planejamento do número de coleta de amostras / The study was conducted in Córrego da Fazenda Glória watershed, Taquaritinga Municipality (SP), in order to assess the spatial variability of physical and chemical attributes of an Ultisol in the area of sugarcane, pasture and native vegetation. Soil sampling was done at the crossing points of a regular grid of 10 m, totaling 110 points in the area of sugarcane, 126 points in the pasture and 88 points in the area of native vegetation. The results of analysis of soil attributes were subjected to analysis of the data, the evaluation of spatial dependence using geostatistics and interpolation of data by kriging to generate maps. The results showed that: a) the soil characteristics studied showed a spatial dependence structure moderate or strong, except for Ca, Mg, SB and V in the area of sugarcane, macroporosity in a pasture area and microporosity porosity, Mg, SB and CTC in the area of native vegetation, b) the physical and chemical were adjusted to exponential and spherical models, and some had no defined structure semivariogram c) the spatial variability of physical and chemical attributes provided visualization of homogeneous management zones, allowing the adoption of a system of precision agriculture, d) native vegetation helps to improve the physical and organic matter in relation to the areas of sugar cane and pasture; e) highest variability measured by the coefficient of variation were found in the area of native vegetation; f) the geostatistical techniques provide a more detailed study of the areas, allowing use in future sampling the range of spatial dependence for planning the number of sampling

Made available in DSpace on 2014-06-11T19:23:24Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-05-29Bitstream added on 2014-06-13T18:50:28Z : No. of bitstreams: 1 pereira_fs_me_jabo.pdf: 218006 bytes, checksum: 0a1a85058e748b7235e5b82d01e0ddc5 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O monitoramento da qualidade do solo pelos atributos físicos é importante para a avaliação da sustentabilidade dos sistemas agrícolas. Um atributo indicador da qualidade do solo deve ser sensível às variações do manejo ao qual está sendo submetido. O intervalo hídrico ótimo (IHO) permite visualizar os efeitos dos sistemas de manejo na melhoria ou na degradação da qualidade física do solo. O índice S é um indicador de qualidade física do solo que indica a distribuição do tamanho dos poros. Objetivou-se com este estudo avaliar a qualidade física do solo utilizando como parâmetros a porosidade, a densidade do solo, o IHO, o índice S, a agregação e a produtividade de milho em um Latossolo Vermelho distrófico, após 11 anos de cultivo de plantas de cobertura em pré-safra ao milho, em sistema de plantio direto com as seguintes plantas de cobertura: crotalária (SDC), milheto (SDM), labe-labe (SDL) e sistema convencional (SC). O delineamento experimental foi o de blocos casualizados, com quatro tratamentos e seis repetições. Os sistemas de plantio direto resultaram em um incremento da porosidade total. O SC não diferiu do SDL e do SDC quanto à macroporosidade. O SC e o SDC apresentaram menor densidade do solo. O intervalo hídrico ótimo nos sistemas de plantio direto foi limitado na parte superior pelo conteúdo de água na capacidade de campo, e na parte inferior pela resistência do solo à penetração. O solo apresentou boa qualidade física, verificado através do IHO. Os sistemas de plantio direto apresentaram maior teor de matéria orgânica em relação ao SC nas camadas de 0-0,05 m. Maiores valores de diâmetro médio ponderado e geométrico dos agregados também foram observados nesse sistema. O solo apresentou boa qualidade física determinada pelo índice S, não apresentando valores inferiores a 0,035. / The monitoring of soil quality by physical attributes is important for the evaluation of the sustainability of agricultural systems. An attribute indicator of soil quality should be sensitive to changes in management which is being submitted. The Least limiting water range (LLWR) will see the effects of management systems or in improving the degradation of soil physical quality. The parameter S is an index of soil physical quality that shows the size distribution of pores. The objective of this study to evaluate the soil physical quality parameters using the porosity, the density of the soil, the LLWR, the index S, the aggregation and productivity in an Oxisol after 11 years of cultivation of plant cover in pre-harvest maize in the conventional system and no tillage with cover plants, sunn hemp (Crotalaria juncea), millet (Pennisetum americanum sin.thiphoydes), lab-lab (Dolichus lablab). The experimental design was a randomized block design with four treatments with split-plot and six replications. The tillage systems resulted in an increase in total porosity. The conventional system did not differ from lablab and sunn hemp on macroporosity. The sunn hemp and the conventional system showed lower bulk density. The least limiting water range of tillage systems has been limited in part by the higher water content at field capacity, and in part by lower resistance to penetration. The soil showed a good physical quality, verified by the LLWR. The tillage systems had higher organic matter content in the layers of the conventional system 0-0.05 m. Higher mean weight diameter of aggregates were also observed in this system. The soil showed a good physical quality determined by the index S, which could not be less than 0.035.