El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / Construction projects for buildings, in the execution stage, show a variation of the real with respect to what is planned in the budget. The consequence of this lag presented usually causes large economic losses. Given this situation, in the construction industry, various control systems have been developed that have as a final deliverable a frequent management report, which serves to make decisions regarding the analyzed, whose results are applied and evaluated after a specific period of time. These management reports have good results at the time of their application; however, data processing, analysis execution, and decision making take approximately 30 days. In this sense, the present investigation proposes to develop a control system focused in real time, generated by the daily control in the execution stage. The results show a reduction of waste in the main resources of multi-family building projects, and this is in labor and materials, identifying the causes that originate them and suggesting generic solutions, at the end of the workday. / Revisión por pares

The intention of this thesis is to develop, evaluate new concepts and look over the current design for a container sized barge module. By request of Group Ocean, a cost and weight reduction is the main improvement criteria along with keeping the strength of the module.Five concepts are developed, analyzed and discussed with the supervisor at Group Ocean, where three are decided to be presented here. The other two are left out, since they are considered way too expensive without giving a satisfying result. The three concepts that are developed throughout this thesis are; changing to high strength steel, changing to sandwich panels and increasing stiffeners with smaller dimensions.A structural optimization is made in the software MATLAB to find out the best dimension to use for the sandwich panels. To determine the local stresses, the finite element method is used in Inventor Professional. It is also where the design and CAD modules are built in, so for simplifications it is used for FEA (Finite Element Analysis) as well. To reduce the amount of elements and nodes, shell elements and other structural constraints are used in the FEA. All the concepts are modelled with the same structural constraints so a practical comparison study can be made.The final designs resulted in a total weight reduction up to 40% with a material cost reduction of 12%. Based on what type of material is chosen, the material cost reduction range is between 3-12% and the weight reduction range is between 13-40%.

Au long des dernières décennies, des importants avancements ont été réalisés dans le domaine de la reconnaissance de la parole à grand vocabulaire. Un des défis à relever dans le domaine concerne la réduction des coûts de développement nécessaires pour construire un nouveau système ou adapter un système existant à une nouvelle tâche, langue ou dialecte. Les systèmes de reconnaissance de la parole à l’état de l’art sont basés sur les principes de l’apprentissage statistique, utilisant l’information fournie par deux modèles stochastiques, un modèle acoustique (MA) et un modèle de langue (ML). Les méthodes standards utilisées pour construire ces modèles s’appuient sur deux hypothèses de base : les jeux de données d’apprentissage sont suffisamment grands, et les données d’apprentissage correspondent bien à la tâche cible. Il est bien connu qu’une partie importante des coûts de développement est dû à la préparation des corpora qui remplissent ces deux conditions, l’origine principale des coûts étant la transcription manuelle des données audio. De plus, pour certaines applications, notamment la reconnaissance des langues et dialectes dits "peu dotés", la collecte des données est en soi une mission difficile. Cette thèse a pour but d’examiner et de proposer des méthodes visant à réduire le besoin de transcriptions manuelles des données audio pour une tâche donnée. Deux axes de recherche ont été suivis. Dans un premier temps, des méthodes d’apprentissage dits "non-supervisées" sont explorées. Leur point commun est l’utilisation des transcriptions audio obtenues automatiquement à l’aide d’un système de reconnaissance existant. Des méthodes non-supervisées sont explorées pour la construction de trois des principales composantes des systèmes de reconnaissance. D’abord, une nouvelle méthode d’apprentissage non-supervisée des MAs est proposée : l’utilisation de plusieurs hypothèses de décodage (au lieu de la meilleure uniquement) conduit à des gains de performance substantiels par rapport à l’approche standard. L’approche non-supervisée est également étendue à l’estimation des paramètres du réseau de neurones (RN) utilisé pour l’extraction d’attributs acoustiques. Cette approche permet la construction des modèles acoustiques d’une façon totalement non-supervisée et conduit à des résultats compétitifs en comparaison avec des RNs estimés de façon supervisée. Finalement, des méthodes non-supervisées sont explorées pour l’estimation des MLs à repli (back-off ) standards et MLs neuronaux. Il est montré que l’apprentissage non-supervisée des MLs conduit à des gains de performance additifs (bien que petits) à ceux obtenus par l’apprentissage non-supervisée des MAs. Dans un deuxième temps, cette thèse propose l’utilisation de l’interpolation de modèles comme une alternative rapide et flexible pour la construction des MAs pour une tâche cible. Les modèles obtenus à partir d’interpolation se montrent plus performants que les modèles de base, notamment ceux estimés à échantillons regroupés ou ceux adaptés à la tâche cible. Il est montré que l’interpolation de modèles est particulièrement utile pour la reconnaissance des dialectes peu dotés. Quand la quantité de données d’apprentissage acoustiques du dialecte ciblé est petite (2 à 3 heures) ou même nulle, l’interpolation des modèles conduit à des gains de performances considérables par rapport aux méthodes standards. / One of the outstanding challenges in large vocabulary automatic speech recognition (ASR) is the reduction of development costs required to build a new recognition system or adapt an existing one to a new task, language or dialect. The state-of-the-art ASR systems are based on the principles of the statistical learning paradigm, using information provided by two stochastic models, an acoustic (AM) and a language (LM) model. The standard methods used to estimate the parameters of such models are founded on two main assumptions : the training data sets are large enough, and the training data match well the target task. It is well-known that a great part of system development costs is due to the construction of corpora that fulfill these requirements. In particular, manually transcribing the audio data is the most expensive and time-consuming endeavor. For some applications, such as the recognition of low resourced languages or dialects, finding and collecting data is also a hard (and expensive) task. As a means to lower the cost required for ASR system development, this thesis proposes and studies methods that aim to alleviate the need for manually transcribing audio data for a given target task. Two axes of research are explored. First, unsupervised training methods are explored in order to build three of the main components of ASR systems : the acoustic model, the multi-layer perceptron (MLP) used to extract acoustic features and the language model. The unsupervised training methods aim to estimate the model parameters using a large amount of automatically (and inaccurately) transcribed audio data, obtained thanks to an existing recognition system. A novel method for unsupervised AM training that copes well with the automatic audio transcripts is proposed : the use of multiple recognition hypotheses (rather than the best one) leads to consistent gains in performance over the standard approach. Unsupervised MLP training is proposed as an alternative to build efficient acoustic models in a fully unsupervised way. Compared to cross-lingual MLPs trained in a supervised manner, the unsupervised MLP leads to competitive performance levels even if trained on only about half of the data amount. Unsupervised LM training approaches are proposed to estimate standard back-off n-gram and neural network language models. It is shown that unsupervised LM training leads to additive gains in performance on top of unsupervised AM training. Second, this thesis proposes the use of model interpolation as a rapid and flexible way to build task specific acoustic models. In reported experiments, models obtained via interpolation outperform the baseline pooled models and equivalent maximum a posteriori (MAP) adapted models. Interpolation proves to be especially useful for low resourced dialect ASR. When only a few (2 to 3 hours) or no acoustic data truly matching the target dialect are available for AM training, model interpolation leads to substantial performance gains compared to the standard training methods.

O presente trabalho teve como objetivo avaliar as características físico-químicas e de aceitação sensorial de diferentes produtos cárneos elaborados com adição de fibra de ervilha como substituto parcial de carne (visando redução de custos no produto) e de gordura (visando a elaboração de um produto mais saudável). Foram processados quatro diferentes produtos cárneos (nuggets, hambúrguer, apresuntado e salsicha), onde cada um apresentou as seguintes formulações: nuggets (Nuggets Controle (NC) - formulação semelhante à comercial, Nuggets Fibra/ Menos Carne (NFMC) - redução de 10% de carne e adição de 2 % de fibra de ervilha e Nuggets Fibra/ Menos Gordura (NFMG) - redução de 10 % de gordura e adição de 2% de fibra de ervilha), hambúrguer (Hambúrguer Controle (HC) - semelhante à formulação comercial; Hambúrguer Fibra/ Menos Carne (HFMC) - redução de 5% de carne e adição de 1 % de fibra de ervilha e Hambúrguer Fibra/ Menos Gordura (FMG) - redução de 7 % de gordura e adição de 1,0 % de fibra de ervilha), apresuntado (Apresuntado Controle (AC) - semelhante à formulação comercial e Apresuntado Fibra/Menos Carne (AFMC) - redução de 5% de carne e adição de 1% de fibra) e salsicha: (Salsicha Controle (SC) - semelhante à formulação comercial; Salsicha Fibra/Menos Carne (SFMC): redução de 5% de carne bovina e adição de 2% de fibra de ervilha e Salsicha Fibra/ Menos Gordura (SFMG) - redução de 7 % de gordura e adição de 2,0 % de fibra de ervilha). Após o processamento os produtos foram caracterizados quanto à composição centesimal, cor objetiva, perfil de textura, perda de peso após a fritura (nuggets), perda de peso por cocção (hambúrguer), redução do diâmetro (hambúrguer), perdas por gotejamento (apresuntado), rendimento do produto (salsicha), estabilidade da emulsão (salsicha) e avaliação sensorial (teste de aceitação com escala hedônica de nove pontos). Para os nuggets as formulações NFMC e NFMG apresentaram pH superior (p<0,05) ao controle, e a formulação NFMG apresentou maior (p<0,05) dureza, quanto à porcentagem de perdas após a fritura, os parâmetros elasticidade e coesividade e a avaliação sensorial não foram encontradas diferenças (p>0,05) entre as formulações. A aplicação de fibra de ervilha e redução de carne e gordura nos hambúrgueres não afetaram (p>0,05) os valor de pH, parâmetros de cor L* e b*, os parâmetros avaliados no perfil de textura, perdas de peso por cocção, redução do tamanho (encolhimento) e avaliação sensorial, apenas a formulação HFMC apresentou redução (p<0,05) no parâmetro a*. Quanto ao apresuntado, os parâmetros de dureza e mastigabilidade foram maiores (p<0,05) para a formulação AC, a formulação AFMC apresentou valor de L* superior (p<0,05), parâmetros de cor a* e b* inferiores (p<0,05) ao C e maior aceitação geral, já o valores de pH, perdas por gotejamento e as médias dos demais atributos sensoriais avaliados não apresentaram diferença (p>0,05) entre as formulações. Em relação aos resultados obtidos na salsicha, a formulação SFMG apresentou pH inferior (p<0,05) a SC e maior valor para os parâmetros de dureza e mastigabilidade. As formulações SFMC e SFMG apresentaram rendimento inferior (p<0,05) a SC e não diferiram quanto à estabilidade da emulsão, já em relação a avaliação sensorial formulação SFMC apresentou notas superiores (p<0,05) à formulação SC e SFMG quanto aos atributos textura, sabor e aceitabilidade geral. De acordo com os resultados obtidos pode-se concluir que a fibra de ervilha mostrou-se como um ingrediente promissor para aplicação em produtos cárneos, já que nas condições testadas a substituição parcial de carne e gordura por fibra de ervilha não ocasionou prejuízo na maioria dos parâmetros físico-químicos avaliados e não prejudicou a aceitação sensorial em nenhum dos produtos cárneos estudados. / The objective of this study was to characterize physicochemical aspects of meat products and their sensorial acceptance elaborated with the addition of pea fiber as a partial substitute of meat (aiming the product cost reduction), and fat (aiming a healthier product). Four different meat products were formulated and processed (chicken nuggets, burger, chopped ham and frankfurter), as follows: chicken nuggets (Nuggets Control (NC) - formulation similar to the commercial product, Nuggets Fiber Less Meat (NFLM) - 10% of meat reduction and 2% of pea fiber addition, and Nuggets Fiber Less Fat (NFLF) - 10% of fat reduction, and 2% of pea fiber addition), burger (Burger Control (HC) - similar to the commercial formulation; Burger Fiber Less Meat (BFLM) - 5% of meat reduction, and 1% of pea fiber addition, and Burger Fiber Less Fat (BFLF) - 7% of fat reduction, and 1% of pea fiber addition); chopped ham (Chopped Ham Control (CHC) - similar to the commercial formulation, and Chopped Ham Fiber Less Meat (CHFLM) - 5% of meat reduction, and 1% of pea fiber addition), chopped ham (Chopped Ham Control (CHC) - similar to the commercial formulation, and Chopped Ham Fiber Less Meat (CHFLM) - 5% of meat reduction, and 1% of pea fiber addition), and frankfurter: ( Frankfurter Control (FC) - similar to the commercial formulation; frankfurter Fiber Less Meat (FFLM): 5% of beef reduction, and 2% pea fiber addition, and Frankfurter Fiber Less Fat (FFLF) - 7% of fat reduction, and 1% of pea fiber addition). After processing the products were characterized concerning proximal composition, color determination, texture, weight loss after frying (chicken nuggets), weight loss after cooking (burger), diameter reduction (burger), drip loss (chopped ham), product outcome (frankfurter), emulsion stability (frankfurter) and sensory analysis (nine point hedonic scale acceptance test). For the nuggets NFLM and NFLF formulations presented higher levels of pH (p<0.05) compared to control, and NFLF formulation presented higher (p<0.05) firmness. Regarding the losses after frying, the elasticity parameters, and cohesiveness and sensory analysis were not different (p>0.05) among formulations. The pea fiber addition and meat and fat reduction in burgers did not affect (p>0.05) the pH levels, color parameters L* and b*; the analyzed parameters concerning texture, weight loss after cooking, diameter reduction (shrinking), and sensory analysis, only BFLM formulation presented reduction (p<0.05) in the parameter a*. Regarding the chopped ham, firmness and chewiness parameters were higher (p<0.05) compared to CHC formulation, CHFLM formulation presented higher L* levels (p<0.05), lower parameters of color a* and b* (p<0.05) compared to CHC and higher general acceptance, yet the pH levels, drip loss and the averages of the other sensory characteristics analyzed did not present difference (p>0.05) among formulations. Regarding the results obtained with the frankfurter, the FFLF formulation presented lower pH levels (p<0.05) compared to FC and higher values for firmness and chewiness parameters, FFLM and FFLF formulations presented lower outcome (p<0.05) compared to FC and were not different to the emulsion stability, however regarding the sensory analysis FFLM formulation presented higher values (p<0.05) compared to FC and FFLF concerning texture, flavor and general acceptance. According to the results it is concluded that pea fiber is a promising ingredient to be added in meat products, since the tested conditions in the partial substitution of meat and fat for pea fiber did not cause significant change in most analyzed physicochemical parameters and did not interfere the sensory acceptance in any of the studied meat products.

Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2015-05-18T11:46:50Z No. of bitstreams: 1 Margarete MenegottoContabeis.pdf: 692048 bytes, checksum: 27f41fb59683bb30748c0b82b5e0b656 (MD5) / Made available in DSpace on 2015-05-18T11:46:50Z (GMT). No. of bitstreams: 1 Margarete MenegottoContabeis.pdf: 692048 bytes, checksum: 27f41fb59683bb30748c0b82b5e0b656 (MD5) Previous issue date: 2010-07-14 / Nenhuma / O presente estudo teve por objetivo identificar quais práticas de gestão econômica e financeira são adotadas por redes de empresas do Rio Grande do Sul. O uso das práticas de gestão econômica e financeira permite aos gestores a avaliação das ações implementadas e a obtenção de informações necessárias para a tomada de decisão. Para a obtenção dos dados sobre a utilização dessas práticas pelas redes de empresas, foi utilizada a técnica de levantamento de dados por meio de entrevistas estruturadas, aplicadas em 21 redes de empresas do estado do Rio Grande do Sul, que compuseram a amostra da pesquisa. O instrumento de coleta de dados abordou aspectos referentes a motivos da formação das redes de empresas; forma de escolha dos associados; estrutura e formalização das redes; práticas de gestão econômica e financeira; governança corporativa; características gerais das redes e perfil dos respondentes. Os dados coletados foram submetidos à análise descritiva, na qual se calculou o Ranking Médio de utilização, bem como percentuais de graus de utilização, ambos baseados nas respostas obtidas na escala Likert. Foi realizada ainda a análise fatorial com intuito de reduzir o número de variáveis e formar fatores, os quais foram submetidos a múltiplas correlações bivariadas no intuito de verificar a relação entre estes. Baseado nessas análises pode-se constatar que as redes de empresas são constituídas com o objetivo principal de aumentar a competitividade, minimizar custos e obter parcerias. Constatou-se que 76% da amostra é formada por associações, e que as redes que possuem volume de transações superior a R$ 10 milhões/ano utilizam as práticas de gestão econômica e financeira com maior intensidade do que as que possuem um volume inferior. Com relação às práticas de gestão econômica e financeira, verificou-se que as cotações de preços com fornecedores potenciais e o fluxo de caixa são as mais utilizadas pela amostra. Os resultados permitem supor, ainda, que quando se trata de gestão econômica e financeira de redes o foco está voltado para o curto prazo, visto que na classificação dos rankings, as três primeiras colocações apresentam esta indicação. / The objective of the present study was to identify which practices of economic and financial management are used by business networks in Rio Grande do Sul. The usage of practices of the economic and financial management allows the evaluation of the actions as well as the obtainment of needed information for decision-making. Structured interviews were used to obtain the data about the usage of these practices.The interviews were applied to 21 business networks in Rio Grande do Sul, which composed the sample of this research. The data collection instrument took into account the aspects referring to the reasons of business network formation; different forms on choosing the associates; structure and formalization of the networks; practices of economic and financial management; corporate governance; general characteristics of networks and respondents’ profile. The data collected was submitted to the descriptive analysis where middle ranking and the percentage of usage were calculated based on the answers obtained on the Likert scale. The factorial analysis was performed in order to reduce the number of variables as well as to form factors which underwent multiple bivariate correlations in order to examine the relationship among them. Based on these analyses, it was possible to verify that business networks are established to increase competitiveness, reduce costs and achieve partnerships. It was found that 76% of the sample is composed of associations, and that the networks that have transaction value exceeding R$ 10 million per year used the practices of economic and financial management with greater intensity than those with less volume. With respectto the practices of economic and financial management, it was found that the price quotation with potential suppliers and cash flow are most used by the sample. The results suggest that, when it comes to economic and financial management of networks, the focus is on short term, since the first three positions have this statement in the rankings classification.

Made available in DSpace on 2016-04-25T18:40:00Z (GMT). No. of bitstreams: 1 Gilmar Pego da Silva.pdf: 1492622 bytes, checksum: 81c819450fee0ff4b286471a0b76a4bc (MD5) Previous issue date: 2014-06-24 / The System Cost Management Target Costing" was developed by the Japanese in the sixties from the technique developed in the United States called Value Engineering, that emerged after World War II due the shortage of raw materials. The "Target Costing" consists in extract the cost from the selling price determined by the market where it s given the maximum allowable cost to reach a desired result (interest). Its application becomes vital as bigger the competitive market and its origins becomes. Even being relatively new (definetivelly has arrived at occident by the 80 s decade and became known by near 2000 s at Brasil) the system was quickly adapted to the local culture, but even so it kept the japanese s fundamentals. The study is hampered because there wasn t any academic connection, what makes the national literature very rare and the same occur to the understanding of cultural factors that induced the adaptations noted in its application. This academic article aimed to collect data and meet through a research that used of descriptive and exploratory method to identify whether the automotive segment automakers affiliated automotive vehicles ANFAVEA (National Association of Automotive Vehicle Makers) use the system management costs "target costing". After tabulating the answers was possible to find out that this kind of industry use, in essence, the system management costs "target costing" and that some adjustments have happened, such as the use of it for new products and also in existing products, focusing the decrease of costs and increasing the profits, even if in some situations the companies accept higher prices than the allowable costs once it has a premise of action to reduce future costs and improve the results. Target costing application in other products act besides direct costs, variable costs, logistic costs and depreciation basis for these costs / O Sistema de Gerenciamento de Custos, denominado Target Costing foi desenvolvido pelos japoneses nos anos sessenta a partir da técnica desenvolvida nos Estados Unidos chamada Engenharia de Valor, que surgiu após a segunda guerra mundial devido a grande escassez de matéria prima. O target costing consiste no cálculo do custo a partir do preço de venda estabelecido pelo mercado onde é administrado o custo máximo admissível para se chegar ao resultado (lucro) desejado. Sua aplicação se torna mais importante quanto maior for o mercado competitivo e a sua origem. Por ser relativamente recente (chegando ao ocidente definitivamente na década de 80 e no Brasil passou a ser divulgado a partir do início dos anos 2000), o sistema foi sofrendo adequações à cultura local, mantendo os princípios criados pelos japoneses. Seu estudo é dificultado, pois não houve uma ligação acadêmica com o tema, o que torna sua literatura nacional escassa, assim como a compreensão dos fatores culturais que causam as adaptações identificadas em sua aplicação. Este trabalho objetivou levantar dados e conhecer através de um questionário utilizando o método de pesquisa exploratória descritiva, identificar se o segmento automotivo de montadoras de veículos automotores filiados a ANFAVEA (Associação Nacional dos Fabricantes de veículos Automotores) utilizam o sistema de gestão de custos target costing. Após a tabulação das respostas da pesquisa foi possível identificar que este segmento utiliza em essência o sistema de gerenciamento de custos target costing e que algumas adequações aconteceram, como a utilização do sistema para novos produtos e também em produtos existentes com foco em redução de custos e aumento de rentabilidade, mesmo que em algumas situações as empresas aceitem custos maiores que o admissível com a premissa de atuação para redução de custo futuro e melhoria de resultados. Sua aplicação em outros produtos vem juntamente com a consideração dos custos diretos, variáveis, custos logísticos e depreciação incremental com base para estes custos

A cana-de-açúcar é uma das mais importantes cadeias do agronegócio brasileiro, além do seu elevado destaque no contexto internacional como fonte de energia renovável e sustentável. Com a competitividade que as Usinas vem passando e pelas perspectivas de crescimento, este trabalho teve como hipóteses: i) o preparo profundo em cana-de-açúcar proporciona um maior volume de solo explorado pelas raízes; ii) o trafégo controlado de máquinas aliado ao espaçamento em fileiras duplas irá trazer uma maior população de plantas, traduzindo isso em incrementos de produtividade e redução de custos para o produtor. O objetivo geral, portanto, foi avaliar métodos de preparo do solo e desenvolver o tráfego contrololado na área, propondo assim que o produtor tenha redução nos custos com o manejo da cultura e também aumente a produtividade e longevidade do canavial. Portanto, pode-se concluir que na análise do comprimento de raízes, em todas as profundidades estudadas, o Preparo Profundo obteve comprimento, área e volume de raízes maiores estatisticamente. A Resistência do solo a penetração no tratamento Preparo Profundo, para as profundidades de 20-40 e 40-60 cm foi de 0,29 MPa e 0,45 MPa, respectivamente. Nas avaliações biométricas da parte área, o Preparo Profundo se destacou com um IAF de 7,1. Os custos das operações no momento do plantio foram bem inferiores no Preparo Profundo, sendo 33% e 58% inferiores aos Mínimo duplo e Convencional duplo, respectivamente. A movimentação do solo causada pelo sistema Preparo Profundo no preparo do solo, foi de até 80% inferior ao tratamento Convencional duplo. O tráfego controlado que o Preparo Profundo proporciona, diminui em até 70% o tráfego de máquinas na lavoura; menor consumo de combustível; redução no tempo das operações agrícolas; redução do parque de máquinas; e menor custo operacional. / Sugarcane is one of the most important crops in the Brazilian agribusiness chain, besides its great branch in the international context as a renewable and sustainable energy source. Through the competitiveness that sugar mills experience and through the growth perspectives of the sector, this study focuses the deep soil preparation for sugarcane crop renewal, since it provides a larger soil volume to be explored by the root system. Furthermore, the machinery traffic controlled by a smaller number of soil preparation operations, reduces farmer´s production costs. The general objective was, therefore, evaluate soil preparation methods and develop a controlled traffic system in the area, in order to propose to the farmer a cost reduction in the farm soil management along with yield increase and extension of the longevity of the crop. As a result of the imposed treatments it was possible to conclude that by the root length analysis in all depths of the Deep Soil Preparation, root length, area and volume were greater. Soil resistance to penetration in the Deep Soil Preparation system was much less in relation to the other treatments, assuming values of 0.29 MPa and 0.45 Mpa in the 20-40 and 40-60 cm layers, respectively. The above ground evaluations showed leaf area index of 7.1 for the Deep Soil Preparation, much higher in relation to the other treatments. Operation costs at planting time were much less for the Deep Soil Preparation, 33% and 58% smaller than for the Double Minimum Tillage and Double conventional, respectively. Soil movement caused by Deep Preparation was up to 80% less in relation to the Double conventional. The controlled machinery traffic that the Deep Soil Preparation provides, reduced up to 70% the farming operations, therefore presenting a lower fuel consumption, a reduction of the time used in farming operations, reduction of the machinery park, and a lower operational cost.

AstraZeneca is one of the largest biopharmaceutical companies in the world, and one of the facilities they have is located in Gärtuna, Södertälje. The facility itself is very big with a floor area of 560.000m2 and has a complex energy system. Caverion holds a facility management contract at AstraZenca, hence operates some of the energy system. The energy investigation of this thesis is part of the work of Caverion to ensure a sustainable energy system in Gärtuna. The energy investigation will include mapping of the energy distribution, seeking for potential of improvements and carry out suggestions for energy optimizations. The methods used during the investigation was a literature study, interviews with personnel of both Caverion and AstraZenca, study of the energy system and calculations relevant to the field of study. The mapping of the energy system includes the heat, steam and cooling distribution. When the mapping of the system was done it was clear that the areas with most potential for improvements were the steam and cooling distribution.The mapping of the steam distribution shows a loss of nearly 46% of the steam at year 2014 and the corresponding cost of about 13,640,000 SEK. Even though the steam distribution showed great potential for improvements, it was found that the work of investigating the system would be too difficult for the scope of the thesis. The cooling distribution however is more accessible and the potential is still high due to low coefficient of performance. Two suggestions for energy optimizations were carried out. The first suggestions involves upgraded electric fan motors for some of the cooling towers, and the second suggestion is to modify existing dry coolers in benefit to utilize free cooling during winter period. The fan motor upgrade based on calculations is estimated to result in a yearly energy saving of at least 1526 MWh and a corresponding cost saving of at least 800,000 SEK per year after the pay-off time (9 months). The dry cooler modification based on calculations is estimated to result in a yearly energy saving of 3053 MWh and a yearly cost saving of 2,083,449 SEK after the pay-off period of 5 months. The investigation carried out in this thesis is relevant to both Caverion and AstraZeneca as it points out the areas with potential of improvements and also gives suggestions on energy optimizations that will reduce energy consumption and result in energy cost savings.

Aggressive scaling required to augment device performance has caused conventional electrode materials to approach their physical scaling limits. Alternative metal gate/high dielectric constant (MG/High-[kappa]) stacks have been implemented successfully in commercial devices and hold promise for further scaling based performance advances. Existing MG/High-[kappa] technology does not achieve a single metal n-MOS to p-MOS effective work function (EWF) tuning range suitable for bulk silicon (Si) device applications. Dual metal gates (DMGs) utilizing a separate metal for n-MOS and p-MOS electrodes increases the cost and complexity of fabrication. The research presented herein introduces a method by which the cost and complexity of MG/High-[kappa] device fabrication may be reduced. Innovative fin field effect transistors (FinFETs) incorporating 3 dimensional ultra thin body silicon on oxide (3-D UTB-SOI) technology display superior electrical characteristics compared to bulk Si devices at the nanometer (nm) dimension and require only a +/-200meV n-MOS to p-MOS EWF tuning range around the Si mid-gap. Single metals capable of achieving this +/-200meV EWF tuning range have been evaluated herein and the tuning mechanisms investigated and engineered to develop a single MG/High-[kappa] FinFET the fabrication complexity of which is reduced by 40%. More specifically, the research shows that the metal thickness of titanium nitride/hafnium silicon oxide (TiN/HfSiOx) gate stack may be engineered to achieve an n-MOS (thinner TiN) to p-MOS (thicker TiN) appropriate FinFET EWF tuning range. FinFETs may be fabricated by depositing a single p-MOS appropriate TiN thickness which may be selectively etched back to achieve thinner, n-MOS appropriate films. Similar electrical behavior is exhibited by etched back and as deposited TiN electrode FinFETs. The single metal etch back fabrication method removes many of the additional steps required for DMG fabrication and preserves the integrity of the MG/High-[kappa] interface between n-MOS and p-MOS devices. These advantages result in reduced fabrication complexity and improved reliability and reproducibility. / text

Water source of life"! Research conducted so far, about life on earth that are proven and tested as we are today would not exist if there was no water. Any form of life depends on water for their survival and for their development; even life forms for more primitive depend on water to survive. Only 2.8%, ie a very low percentage of the universe on the Water Planet Earth is sweet and much of it is unfit for human consumption and requires the use of high technology to make water potable. In this setting, unless we take immediate action we will not have the liquid of life, and following this thought and the ways proposed in the meeting of 92 countries in Rio, in its action plan for the 21st century, aiming at the sustainability of life on Earth, in one of its 40 chapters, Agenda 21 addresses the "conservation and management of natural resources." Therefore, the adoption of measures that will qualify for the management and promote changes in the perception of how to work in relation to impacts on catchment and natural resource conservation, and "Think Globally and Act Locally" (Agenda 21) is the purpose this work. In this scenario applies today the polluter-pays principle. Thus the actual charge to the water, which is a high value, encourages water rationing benefit while minimizing the impacts of their scarcity. Innovations in saw the need to establish standards for the construction of toilets and other equipment that would reduce consumption. But that alone is not enough, it is necessary to promote alternatives to find a solution for use in a home rainwater on items that do not require investment in benefit it. It follows therefore that the various solutions over the years has been researched and proven to lower investment cost and installation of a system to capture rainwater, which allows an economy of treated water in more that 41% only use in toilets according to studies conducted in the Netherlands. The analysis of these facts justify the proposal of the work, which is looking for innovations result from a millennial issue, especially for the process of cutting costs, reducing the use of natural resources as: "Use of Rainwater without treatment in a Residential ". / Água fonte da vida ! Pesquisas realizadas, até o momento, sobre a vida na terra são provadas e comprovadas que não existiríamos como somos hoje se não houvesse água. Qualquer forma de vida depende da água para a sua sobrevivência e para o seu desenvolvimento; mesmo formas de vidas por mais primitivas dependem da água para sobreviver. Somente 2,8%, ou seja, um percentual muito baixo diante do universo da Água do Planeta Terra é doce e grande parte desta é imprópria para o consumo humano sendo necessário o emprego de alta tecnologia para tornar a água potável. Diante deste cenário, se não tomarmos uma atitude imediata não teremos mais o líquido da vida, e seguindo este pensamento e os caminhos propostas no encontro dos países na Rio 92, em seu plano de ação para o século 21, visando a sustentabilidade da Vida na Terra, em um dos seus 40 capítulos, a Agenda 21 aborda a conservação e manejo dos recursos naturais . Portanto, a adoção de medidas para que venham qualificar a gestão e promover alterações na percepção quanto à forma de trabalhar em relação aos impactos causados na captação e conservação dos recursos naturais, e Pensar Globalmente e Agir Localmente (Agenda 21) é o propósito deste trabalho. Diante deste cenário aplica-se hoje o princípio do poluidor-pagador. Com isso a própria cobrança realizada sobre a água, que é um valor elevado, incentiva o racionamento da água beneficiada, minimizando os impactos da sua escassez. Nas inovações viu-se a necessidade de estabelecer normas para a construção de bacias sanitárias assim como outros equipamentos que reduzissem o consumo. Mas somente isso não basta, é necessário à promoção de alternativas em busca de uma solução para uso em uma residência da água da chuva em itens que não necessitem o investimento em beneficiá-la. Conclui-se, portanto que das diversas soluções que ao longo dos anos vem sendo pesquisado e comprovadamente a de menor custo de investimento e a instalação de um sistema de captação de água de chuva, qual possibilita uma economia de água tratada em mais de 41% somente na utilização em vasos sanitários conforme estudos realizados na Holanda. As análises destes fatos justificam a presente proposta do trabalho, que tem por resultado buscar inovações de um assunto milenar, principalmente, para o processo de redução de custos, redução do uso dos Recursos Naturais como; Aproveitamento de Água de Chuva sem Tratamento em uma Residencial .