A surface reconstruction and image enhancement non-linear finite element technique based on minimization of L1 norm of the total variation of the gradient is introduced. Since minimization in the L1 norm is computationally expensive, we seek to improve the performance of this algorithm in two fronts: first, local L1- minimization, which allows parallel implementation; second, application of the Augmented Lagrangian method to solve the minimization problem. We show that local solution of the minimization problem is feasible. Furthermore, the Augmented Lagrangian method can successfully be used to solve the L1 minimization problem. This result is expected to be useful for improving algorithms computing digital elevation maps for natural and urban terrain, fitting surfaces to point-cloud data, and image super-resolution.

The electric power sector is undergoing significant changes with the development of Smart Grid technologies and is rapidly influencing the way we consume energy. Demand Response (DR) is an important element in the emerging smart grid paradigm and is paving way for the more sophisticated implementation of Energy Hub Management Systems (EHMSs). Utilities are looking at Demand Side Management (DSM) and DR services that allow customers to make informed decisions regarding their energy consumption which in return, can help the energy providers to reduce their peak demand and hence enhance grid sustainability. Ice rinks are large commercial buildings which facilitate various activities such as hockey, figure skating, curling, recreational skating, public arenas, auditoriums and coliseums. These have a complex energy system; in which an enormous sheet of ice is maintained at a low temperature while at the same time the spectator stands are heated to ensure comfortable conditions for the spectators. Since indoor ice rinks account for a significant share of the commercial sector and are in operation for more than 8 months a year, their contribution in the total demand cannot be ignored. There is significant scope for energy savings in indoor ice rinks through optimal operation of their climate control systems. In this work, a mathematical model of indoor ice rinks for the implementation of EHMS is developed. The model incorporates weather forecast, electricity price information and end-user preferences as inputs and the objective is to shift the operation of climate control devices to the low electricity price periods, satisfying their operational constraints while having minimum impact on spectator comfort. The inside temperature and humidity dynamics of the spectator area are modeled to reduce total electrical energy costs while capturing the effect of climate control systems including radiant heating system, ventilation system and dehumidification system. Two different pricing schemes, Real Time Pricing (RTP) and Time-of-Use (TOU), are used to assess the model, and the resulting energy costs savings are compared. The expected energy cost savings are evaluated for a 8 month period of operation of the rink incorporating the uncertainties in electricity price, weather conditions and spectator schedules through Monte Carlo simulations. The proposed work can be implemented as a supervisory control in existing climate controllers of indoor ice rinks and would play a significant role in the enforcement of EHMS in Smart Grids.

Numerical simulation is an important tool in understanding the electromagnetic field and how it interacts with the environment. Different topics for time-domain finite-difference (FDTD) and finite-element (FETD) methods for Maxwell's equations are treated in this thesis. Subcell models are of vital importance for the efficient modeling of small objects that are not resolved by the grid. A novel model for thin sheets using shell elements is proposed. This approach has the advantage of taking into account discontinuities in the normal component of the electric field, unlike previous models based on impedance boundary conditions (IBCs). Several results are presented to illustrate the capabilities of the shell element approach. Waveguides are of fundamental importance in many microwave applications, for example in antenna feeds. The key issues of excitation and truncation of waveguides are addressed. A complex frequency shifted form of the uniaxial perfectly matched layer (UPML) absorbing boundary condition (ABC) in FETD is developed. Prism elements are used to promote automatic grid generation and enhance the performance. Results are presented where reflection errors below -70dB are obtained for different types of waveguides, including inhomogeneous cases. Excitation and analysis via the scattering parameters are achieved using waveguide modes computed by a general frequency-domain mode solver for the vector Helmholtz equation. Huygens surfaces are used in both FDTD and FETD for excitation in waveguide ports. Inverse problems have received an increased interest due to the availability of powerful computers. An important application is non-destructive evaluation of material. A time-domain, minimization approach is presented where exact gradients are computed using the adjoint problem. The approach is applied to a general form of Maxwell's equations including dispersive media and UPML. Successful reconstruction examples are presented both using synthetic and experimental measurement data. Parameter reduction of complex geometries using simplified models is an interesting topic that leads to an inverse problem. Gradients for subcell parameters are derived and a successful reconstruction example is presented for a combined dielectric sheet and slot geometry.

Η αλματώδης ανάπτυξη που παρατηρείται τα τελευταία χρόνια στον τομέα των οπτικών δικτύων τηλεπικοινωνιών, έχει ως συνέπεια την αύξηση της κατανάλωσης ενέργειας. Σήμερα υπολογίζεται ότι η ενέργεια που οφείλεται σε IT τεχνολογίες αποτελεί το 1%-2% της παγκόσμιας κατανάλωσης. Ωστόσο, τα ποσοστά αυτά αναμένεται να αυξηθούν δραματικά μιας και όλο και περισσότερες εφαρμογές, που απαιτούν μεγάλο εύρος ζώνης (π.χ. video streaming, HDTV), αναπτύσσονται ταχύτατα. Μέχρι πρόσφατα ο κύριος περιορισμός στην ανάπτυξη των δικτύων ήταν το κόστος εξοπλισμού και μετάδοσης. Σήμερα το μεγαλύτερο πρόβλημα είναι η όλο και αυξανόμενη κατανάλωση ενέργειας. Έτσι λοιπόν πιο energy-aware προσεγγίσεις για το σχεδιασμό, την υλοποίηση και τη λειτουργία των οπτικών δικτύων κρίνονται πλέον απαραίτητες. Για την αντιμετώπιση του προβλήματος της κατανάλωσης ενέργειας σε οπτικά δίκτυα υπάρχουν δυο προσεγγίσεις: η ανάπτυξη ενεργο-αποδοτικών συσκευών δικτύου ή η ανάπτυξη αλγορίθμων δρομολόγησης και ανάθεσης μήκους κύματος (RWA) που να λαμβάνουν υπόψη τους και την ενέργεια που καταναλώνει το δίκτυο. Στην παρούσα διπλωματική εργασία επιλέχθηκε η δεύτερη προσέγγιση όπου και αναπτύχθηκαν Energy Aware RWA αλγόριθμοι για τον σχεδιασμό WDM οπτικών δικτύων με στόχο την ελαχιστοποίηση της ενέργειας που καταναλώνεται. Αυτό επιτυγχάνεται μειώνοντας τον αριθμό των συσκευών του δικτύου, όπως αναγεννητές, ενισχυτές, add/drop τερματικά κ.τ.λ. που είναι γνωστό ότι καταναλώνουν αρκετή ενέργεια κατά τη λειτουργία τους. Αρχικά αναπτύχθηκε ένας αλγόριθμος σε Ακέραιο Γραμμικό Ακέραιο Προγραμματισμό (Integer Linear Programming - ILP) που ελαχιστοποιεί από κοινού όλες εκείνες τις συσκευές του δικτύου που καταναλώνουν ενέργεια. Έπειτα το πρόβλημα χωρίζεται σε δυο υπο-προβλήματα: την κατανομή αναγεννητών σε κόμβους του δικτύου όπου η αναγέννηση είναι απαραίτητη και στο Energy Aware RWA. Σε αυτή τη περίπτωση εκτός από ILP προγραμματισμό εφαρμόζεται επίσης και χαλάρωση των περιορισμών ακεραιότητας (LP προγραμματισμός) ώστε να μπορούν να προκύψουν λύσεις και για μεγαλύτερες τοπολογίες. Για τον υπολογισμό της απόδοσης των προτεινόμενων αλγορίθμων διεξάγονται μια σειρά από πειράματα εξομοιώσεων σε Matlab, με τη βοήθεια του LINDO API για την επίλυση των (I)LP προβλημάτων σε ένα μικρό δίκτυο αλλά και στο ευρωπαϊκό δίκτυο Geant-2. Oι προτεινόμενοι αλγόριθμοι συγκρίνονται με τυπικούς RWA αλγορίθμους που δεν λαμβάνουν υπόψη την ενέργεια ή ελαχιστοποιούν κάποιες από τις συνιστώσες του δικτύου. Οι υπολογισμοί δείχνουν ότι η κατανάλωση της ενέργειας του δικτύου μπορεί να μειωθεί μέχρι και 45% σε δίκτυο με συνδέσμους μονής ίνας και μέχρι και 65% σε συνδέσμους διπλής ίνας. / The continuing deployment and upgrade of optical telecommunication networks drive up power and energy consumption, in a way that makes operators worry that future energy consumption levels may pose constraints on communications growth that are more significant than those posed by bandwidth considerations. As community concerns about global energy consumption grow, the power consumption of the optical networks is becoming an issue of increasing importance. Even though energy consumption of IT-related infrastructure currently represents between 1% and 2% of global energy consumption, such values are destined to rise, as more bandwidth-hungry applications (such as video streaming, and HDTV in particular) continue to be developed. Thus, it seems that an energy-aware approach is increasingly needed during the design, implementation and operation of optical networks. Two different approaches can be explored to reduce energy consumption in optical networks: the improvement of the energy efficiency of the equipment and the energy awareness of the Routing and Wavelengths Assignment (RWA) algorithms. In this work the energy minimization problem in optical networks is considered from an algorithmic perspective. The objective of the proposed algorithms is to plan optical WDM networks so as to minimize the energy consumed, by minimizing the number of the most energy-consuming components. Such components can be amplifiers, regenerators, add/drop terminals, optical fibers, etc. Firstly an Integer Linear Programming (ILP) formulation is given that aims at jointly minimizing the energy consuming modules present in an optical network, such as the regenerators, amplifiers, wavelength selective switches, etc. Then the problem for a general (translucent network) is decomposed into a regeneration placement problem, and an energy-aware RWA for transparent networks problem, where again ILP formulations are given to minimize energy consumption. The ILP of the decomposed problem is solved as an LP problem, by relaxing the integer constraints, in order to obtain solutions for larger network topologies. In order to evaluate the performance of the proposed Energy Aware (EA)-RWA algorithms a number of simulation experiments is carried out. All the RWA algorithms are implemented in Matlab and LINDO API is used to solve the corresponding (I)LP problems. The proposed algorithms are compared with RWA algorithms that do not consider energy minimization techniques or RWA algorithms that partially take into account energy consuming components. The simulation results showed that the energy consumption in the optical layer can be reduced by up to 45% in single fiber networks and by up to 65% in multi-fiber networks.

Thesis (MSc)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: As the world enters a new millennium, global awareness of environmental issues have reached new heights. No longer is the environment seen as an all-absorbing and allproviding resource. Rather, there exists a greater awareness that without active management, the inherent risks and hazards - posed by both the natural environment and humankind's interaction with this environment - will be realised. In order to ensure a sustainable co-existence, humanity's interaction with the surrounding. environment must be managed in a responsible manner. In this regard, all aspects of this interaction require attention. Various frameworks, manifested in a variety of forms, have been proposed. One of the most basic aspects of species existence is the generation of waste. Human existence is no different. However, as a result of the variety of activities that characterise humans' existence on Earth, the generation of waste represents the most tangible and probably the most threatening aspect of this interaction. This study focuses on waste management from an environmental perspective at a specific institution, namely the University of Stellenbosch. In terms of its findings, the study has established the applicability of various regulatory and institutional frameworks within which the University operate. These frameworks address the role of the University both in terms of waste management and sustainable development. It has established the types of waste generated, the generating processes, the storage, and removal and disposal characteristics of waste management at the University. Waste management at the University is determined to be complex and fragmented, the result of a variety of activities occurring. It has established that limited waste minimisation and reduction activities occur, even though the University had identified effective and efficient resource use as a strategic priority. Although attempts at improving waste management are admittedly underway, the study has found that adopting an environmental management system approach to waste management will enable the University to meet pending legislative and institutional environmental commitments. / AFRIKAANSE OPSOMMING: Met die binnetree van 'n nuwe millennium het wêreldwye bewustheid van omgewingskwessies nuwe hoogtes bereik. Die omgewing word nie meer bloot beskou as 'n alles-absorberende en alles-voorsienende hulpbron nie. Veel eerder groei bewustheid van die noodsaak vir aktiewe bestuursingryping om die gevare en bedreigings inherent aan die natuurlike omgewing, en die mens se interaksie daarmee, die hoof te bied. Ter wille van volhoubare saambestaan moet die mens-omgewing interaksie op verantwoordelike wyse bestuur word. Alle aspekte van hierdie interaksie vereis aandag en 'n verskeidenheid raamwerke in 'n verskeidenheid formate is hiervoor voorgestel. Die produksie van afval is een van die mees basiese kenmerke van lewensbestaan. Menslike bestaan is nie daarbo verhewe nie. Weens die verskeidenheid aktiwiteite wat menslike bestaan kenmerk, is die produksie van afval die mees tasbare en waarskynlik mees bedreigende manifestasie van mens-omgewing interaksie. Hierdie tesis fokus op afvalbestuur uit die invalshoek van die omgewing, soos dit beslag kry aan 'n spesifieke instansie, naamlik die Universiteit van Stellenbosch. Die studie het ten aanvang die implikasies van die verskeidenheid regulatoriese en institusionele raamwerke waarin die universiteit opereer, bevestig. Hierdie raamwerke omvat die rol van die Universiteit beide in terme van afvalbestuur en volhoubare ontwikkeling. Die tipes afval wat aan die instansie gegenereer word is geklassifiseer en die genererings-, opbergings-, verwyderings- en verwerkingsprosesse en verantwoordelikhede is vasgestel en gedokumenteer. Afvalbestuur aan die Universiteit blyk kompleks en gefragmenteerd te wees, hoofsaaklik as gevolg van die wye verskeidenheid afval-skeppingsbronne wat hier aangetref word. Ook is bevind dat, ten spyte van die instansie se identifisering van effektiewe hulpbrongebruik as 'n strategiese prioriteit, slegs beperkte afval-inkorting en -reduksie aan die instansie plaasvind. Hoewel daar tans 'n aktiewe proses geloods word om afvalbestuur te verbeter, beveel die studie die aanvaarding, ontwerp en implementering van 'n veel meer wydlopende en holistiese benadering in die vorm van 'n geïntegreerde omgewingsbestuurstelsel aan. Slegs hierdeur sal die Universiteit in staat wees om aan die volgende vlaag wetgewing rakende afvalbestuur te voldoen en ook verantwoordelike omgewingsverbintenis te demonstreer.

Il y a un intérêt grandissant pour le développement d’outils de traitements adaptés aux images multimodales (plusieurs images de la même scène acquises avec différentes caractéristiques). Permettant une représentation plus complète de la scène, ces images multimodales ont de l'intérêt dans plusieurs domaines du traitement d'images, mais les exploiter et les manipuler de manière optimale soulève plusieurs questions. Cette thèse étend les représentations hiérarchiques, outil puissant pour le traitement et l’analyse d’images classiques, aux images multimodales afin de mieux exploiter l’information additionnelle apportée par la multimodalité et améliorer les techniques classiques de traitement d’images. Cette thèse se concentre sur trois différentes multimodalités fréquemment rencontrées dans le domaine de la télédétection. Nous examinons premièrement l’information spectrale-spatiale des images hyperspectrales. Une construction et un traitement adaptés de la représentation hiérarchique nous permettent de produire une carte de segmentation de l'image optimale vis-à-vis de l'opération de démélange spectrale. Nous nous concentrons ensuite sur la multimodalité temporelle, traitant des séquences d’images hyperspectrales. En utilisant les représentations hiérarchiques des différentes images de la séquence, nous proposons une nouvelle méthode pour effectuer du suivi d’objet et l’appliquons au suivi de nuages de gaz chimique dans des séquences d’images hyperspectrales dans le domaine thermique infrarouge. Finalement, nous étudions la multimodalité sensorielle, c’est-à-dire les images acquises par différents capteurs. Nous appuyant sur le concept des tresses de partitions, nous proposons une nouvelle méthodologie de segmentation se basant sur un cadre de minimisation d’énergie. / There is a growing interest in the development of adapted processing tools for multimodal images (several images acquired over the same scene with different characteristics). Allowing a more complete description of the scene, multimodal images are of interest in various image processing fields, but their optimal handling and exploitation raise several issues. This thesis extends hierarchical representations, a powerful tool for classical image analysis and processing, to multimodal images in order to better exploit the additional information brought by the multimodality and improve classical image processing techniques. %when applied to real applications. This thesis focuses on three different multimodalities frequently encountered in the remote sensing field. We first investigate the spectral-spatial information of hyperspectral images. Based on an adapted construction and processing of the hierarchical representation, we derive a segmentation which is optimal with respect to the spectral unmixing operation. We then focus on the temporal multimodality and sequences of hyperspectral images. Using the hierarchical representation of the frames in the sequence, we propose a new method to achieve object tracking and apply it to chemical gas plume tracking in thermal infrared hyperspectral video sequences. Finally, we study the sensorial multimodality, being images acquired with different sensors. Relying on the concept of braids of partitions, we propose a novel methodology of image segmentation, based on an energetic minimization framework.

Gradient descent methods and especially their stochastic variants have become highly popular in the last decade due to their efficiency on big data optimization problems. In this thesis we present the development of data sampling strategies for these methods. In the first four chapters we focus on four views on the sampling for convex problems, developing and analyzing new state-of-the-art methods using non-standard data sampling strategies. Finally, in the last chapter we present a more flexible framework, which generalizes to more problems as well as more sampling rules. In the first chapter we propose an adaptive variant of stochastic dual coordinate ascent (SDCA) for solving the regularized empirical risk minimization (ERM) problem. Our modification consists in allowing the method to adaptively change the probability distribution over the dual variables throughout the iterative process. AdaSDCA achieves a provably better complexity bound than SDCA with the best fixed probability distribution, known as importance sampling. However, it is of a theoretical character as it is expensive to implement. We also propose AdaSDCA+: a practical variant which in our experiments outperforms existing non-adaptive methods. In the second chapter we extend the dual-free analysis of SDCA, to arbitrary mini-batching schemes. Our method is able to better utilize the information in the data defining the ERM problem. For convex loss functions, our complexity results match those of QUARTZ, which is a primal-dual method also allowing for arbitrary mini-batching schemes. The advantage of a dual-free analysis comes from the fact that it guarantees convergence even for non-convex loss functions, as long as the average loss is convex. We illustrate through experiments the utility of being able to design arbitrary mini-batching schemes. In the third chapter we study importance sampling of minibatches. Minibatching is a well studied and highly popular technique in supervised learning, used by practitioners due to its ability to accelerate training through better utilization of parallel processing power and reduction of stochastic variance. Another popular technique is importance sampling { a strategy for preferential sampling of more important examples also capable of accelerating the training process. However, despite considerable effort by the community in these areas, and due to the inherent technical difficulty of the problem, there is no existing work combining the power of importance sampling with the strength of minibatching. In this chapter we propose the first importance sampling for minibatches and give simple and rigorous complexity analysis of its performance. We illustrate on synthetic problems that for training data of certain properties, our sampling can lead to several orders of magnitude improvement in training time. We then test the new sampling on several popular datasets, and show that the improvement can reach an order of magnitude. In the fourth chapter we ask whether randomized coordinate descent (RCD) methods should be applied to the ERM problem or rather to its dual. When the number of examples (n) is much larger than the number of features (d), a common strategy is to apply RCD to the dual problem. On the other hand, when the number of features is much larger than the number of examples, it makes sense to apply RCD directly to the primal problem. In this paper we provide the first joint study of these two approaches when applied to L2-regularized ERM. First, we show through a rigorous analysis that for dense data, the above intuition is precisely correct. However, we find that for sparse and structured data, primal RCD can significantly outperform dual RCD even if d ≪ n, and vice versa, dual RCD can be much faster than primal RCD even if n ≫ d. Moreover, we show that, surprisingly, a single sampling strategy minimizes both the (bound on the) number of iterations and the overall expected complexity of RCD. Note that the latter complexity measure also takes into account the average cost of the iterations, which depends on the structure and sparsity of the data, and on the sampling strategy employed. We confirm our theoretical predictions using extensive experiments with both synthetic and real data sets. In the last chapter we introduce two novel generalizations of the theory for gradient descent type methods in the proximal setting. Firstly, we introduce the proportion function, which we further use to analyze all the known block-selection rules for coordinate descent methods under a single framework. This framework includes randomized methods with uniform, non-uniform or even adaptive sampling strategies, as well as deterministic methods with batch, greedy or cyclic selection rules. We additionally introduce a novel block selection technique called greedy minibatches, for which we provide competitive convergence guarantees. Secondly, the whole theory of strongly-convex optimization was recently generalized to a specific class of non-convex functions satisfying the so-called Polyak- Lojasiewicz condition. To mirror this generalization in the weakly convex case, we introduce the Weak Polyak- Lojasiewicz condition, using which we give global convergence guarantees for a class of non-convex functions previously not considered in theory. Additionally, we give local convergence guarantees for an even larger class of non-convex functions satisfying only a certain smoothness assumption. By combining the two above mentioned generalizations we recover the state-of-the-art convergence guarantees for a large class of previously known methods and setups as special cases of our framework. Also, we provide new guarantees for many previously not considered combinations of methods and setups, as well as a huge class of novel non-convex objectives. The flexibility of our approach offers a lot of potential for future research, as any new block selection procedure will have a convergence guarantee for all objectives considered in our framework, while any new objective analyzed under our approach will have a whole fleet of block selection rules with convergence guarantees readily available.

Objetivo: Comparar as relações de custo-efetividade de duas terapias, Imunoglubulina Intravenosa (IgIV) e Plasmaferese (PE), no tratamento da Síndrome de Guillain-Barré sob a perspectiva do sistema público (SUS). O objetivo secundário foi avaliar a adesão às recomendações da Comissão de Medicamentos do HCPA Métodos: estudo transversal com análise econômica de pacientes tratados por Síndrome de Guillain-Barré no período de junho de 2003 a junho de 2008 no Hospital de Clínicas de Porto Alegre (HCPA). Foi realizada análise de custo-efetividade do emprego de IgIV e de PE nestes pacientes, pelo método de minimização de custos, considerando-se somente os custos diretos sanitários, fornecidos pelo sistema gerencial da instituição . Foram excluídos os pacientes que usaram outro tipo de tratamento associado ou isolado. Coletaram-se os dados através da revisão dos prontuários. A gravidade da doença na internação foi classificada como: doença leve, quando caminhar foi possível; doença moderada, quando caminhar foi impossível; doença grave, quando os pacientes necessitaram de ventilação assistida. A incapacidade na alta foi estabelecida pela escala de sete pontos de Hughes. A adesão às recomendações da Comissão de Medicamentos do HCPA, objetivo secundário, foi avaliada através da dose e o esquema de prescrição da IgIV. Resultados: Vinte e cinco participantes (2 a 70 anos) foram incluídos no estudo, cinco tratados com PE, empregando-se Albumina Humana como substituto do plasma, e 20 tratados com IgIV. O custo total do tratamento de um paciente com PE foi R$10.603,88 (± 2.978,12) e o de um que recebeu IgIV foi R$ 32.103,00 (± 21.454,24). O custo total da internação foi de R$45.027,14 (± 32.750,45) para os tratados com PE e de R$ 60.844,28 (±48.590,52) para os que receberam IgIV. Em relação ao desfecho clínico principal, melhora na escala de incapacidade de sete pontos, após o tratamento com uma das alternativas escolhida, a mediana dos pacientes que internaram com grau de gravidade 3 e que foram tratados com PE foi igual a dos que receberam IgIV. Em relação à permanência hospitalar, permanência em UTI e dias de Ventilação Mecânica, não houve diferença estatisticamente significativa entre os dois tratamentos. Conclusões: Quando comparados os custos médios das duas opções terapêuticas, uma delas aparece claramente com menor custo. Quando comparados os desfechos, após o emprego de cada opção terapêutica, estes não revelam diferença. Concluímos que, no HCPA, a opção pelo procedimento Plasmaferese é mais custo efetiva do que o emprego da IgIV. / Objectives: To compare the cost-effectiveness of two distinct therapies, Intravenous Immunoglobulin (IVIg) and Plasma Exchange (PE) in the treatment of Guillain-Barré Syndrome, concerning the public health care system. Compliance to the guidelines of the Pharmacy and Therapeutics Committee of the Hospital de Clínicas de Porto Alegre was a secondary objective. Methods: A cross-sectional, economical analysis was conducted, including patients treated for GBS in the period from June, 2003 through June, 2008 in Hospital de Clínicas de Porto Alegre (HCPA). The cost-effectiveness of the use of IVIg and PE in such patients was studied through the cost minimization method, considering direct medical costs only (2008 currency), yield by the management of the institution. Patients receiving treatments other than PE or IVIg were excluded. Data were collected by chart reviews. Severity of disease on admittance was classified as follows: mild disease, when the patient was able to walk; moderate disease, when the patient was unable to walk, and severe disease, when assisted ventilation was required. Disability on discharge was established by the 7-point scale of Hughes. Compliance to the guidelines of the Pharmacy and Therapeutics Committee was evaluated through the dose and prescription scheme of IVIg. Results: Twenty-five participants (2 to 70 years of age) were included in the study, 5 were submitted to treatment with PE, using human albumin as replacement for plasma, and 20 were treated with IVIg. The total treatment cost for PE in a single patient was US$6,058.85 (±1,701.78 SD), and the same expense for IVIg was US$18,344.57 (± 12,259.56 SD) (p = 0.035). Total inpatient cost was US$25,729.79 (± 18,714.54 SD) in the PE group, and US$34,768.16 (±27,766.01 SD) (p=0.530) in the IVIg group. The main clinical outcome was improvement in the 7-point disability grade scale. The median of that measure in patients admitted with a severity grade 3 treated either with PE and IVIg was the same. Secondary outcomes, such as in-hospital stay, ICU stay, and number of days on mechanical ventilation revealed no statistically significant difference between treatments. Conclusions: As the mean expenses of both therapeutic options are compared, one clearly stands-out as less onerous. Clinical outcomes, when compared, reveal no statistical difference after each treatment. We concluded that, in HCPA, plasma exchange is more cost-effective than intravenous immunoglobulin.

Este estudo tem como objetivo avaliar os sistemas de gestão ambiental em cinco indústrias petroquímicas no Estado do Rio de Janeiro, com foco no gerenciamento de resíduos sólidos e líquidos, gerados no processo produtivo, considerando-se a legislação brasileira aplicada ao meio ambiente. Acompanhando a evolução do sistema de gestão ambiental adotado pelos diferentes segmentos produtivos, observa-se cada vez mais a importância da adequação do processo a condições que permitam a sustentabilidade ambiental. As indústrias em questão são consideradas como referência no setror petroquímico de transformação, com atividades altamente impactantes ao meio ambiente. O objetivo geral deste estudo é conhecer a metodologia empregada por cada uma destas indústrias, na busca da melhoria contínua de gestão, face aos desafios de utilização de recursos renováveis e não renováveis, tratamentos de seus resíduos sólidos e líquidos, em compatibilidade com melhores e mais eficientes resultados de produção, bem como, a sustentabilidade do negócio. Para o desenvolvimento deste trabalho procedeu-se à Pesquisa Aplicada e em seguida ao Estudo de Caso. Primeiramente foram realizadas visitas às indústrias, utilizando-se um questionário para avaliar questões relativas ao gerenciamento de resíduos sólidos e líquidos do processo produtivo. Como Estudo de Caso selecionou-se uma das indústrias, que não possui todas as certificações relativas às normas ambientais e também não apresenta estação de tratamento de água e de efluentes na sua unidade industrial. Esta empresa foi comparada ás demais como referência para a identificação das iniciativas que as indústrias petroquímicas vem adotando, relacionadas com a preservação dos recursos naturais e minimização da geração de resíduos. Esta análise comparativa permitiu incluir neste estudo recomendações para implementação ou melhoria em aspectos do processo considerados impactantes ao meio ambiente e que, porventura, ainda não se encontravam em conformidade com atendimento às leis ambientais, e/ou, que não tenham atingido um nível de minimização de resíduos de forma a promover sua sustentabilidade ambiental. / This research has the aim of evaluating environmental management systems of five petrochemical industries in the State of Rio de Janeiro. This studys main focus is on the management of solid and liquid wastes which are generated during producing processes and the commitment with Brazilian laws applied to environmental protection. Following the development on environmental management systems used by different producing market segments, it is noticed that more and more importance is given to adapt a production process to limits and conditions that allows its environmental sustainability. The five above-mentioned studied industries are reference in the transformation sector, with productive activities which may cause critical impacts on the environment. The general target of this study is to give a closer look on the methodology used by each of those industries on their efforts for continuous management improvement, facing the challenges of the use of renewable and not renewable resources, solid and liquid waste treatment, in agreement with more efficient production results, as well as the business sustainability. For the development of this study it was divided in two parts: Applied Research and Case Study. First, the industries were visited and the questionnaire was applied to evaluate issues concerning management of solid and liquid waste generated during the process. One of the five industries was selected as Case Sudy, once it does not have all the certifications complying environmental norms and also it does not have wastewater treatment at its facility. This industry was compared to the others as a reference to identify the efforts that are being followed by petrochemical industries, concerning natural resources preservation and residues minimization. This comparative analysis allowed recommendations to improve process steps considered highly harmful to the environment, and, by the way are not still accomplishing to Brazilian environmental laws, and/or that have not yet achieved an accepted waste minimization level in order to assure its environmental sustainability.

Submitted by Maike Costa (maiksebas@gmail.com) on 2016-03-31T13:04:57Z No. of bitstreams: 1 arquivo total.pdf: 1303363 bytes, checksum: ab5b7fb3a9fd956428d4b499afd3b94c (MD5) / Made available in DSpace on 2016-03-31T13:04:57Z (GMT). No. of bitstreams: 1 arquivo total.pdf: 1303363 bytes, checksum: ab5b7fb3a9fd956428d4b499afd3b94c (MD5) Previous issue date: 2015-10-14 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / In this work we present an abstract version of the concentration compactness principle by Lions, extending it to Hilbert spaces. To do so, we include the concept of dislocation space, the pair (H;D) formed by a separable Hilbert space H (being H1(RN) the model case, N 3) and a set D of linear limited operators on H; as well as the concept of the D-weak convergence. The main result of this theory is, in a sense, a generalization of the famous theorem of Banach-Bourbaki-Alaoglu. Another important consequence of the theory is the equivalence of D-weak convergence in H1(RN);N 3 and strong convergence in Lp; for p 2 (2; 2 ) and D appropriate. With this version, we prove existence of solution for some classes of elliptic problem on unbounded domains, via constrained minimization method. / Neste trabalho apresentamos uma versão abstrata do princípio de concentração de compacidade de Lions, estendo-o para espaços de Hilbert. Para tanto, incluímos o conceito de espaço de deslocamento, o par (H;D); formado por um espaço de Hilbert H separável (sendo H1(RN) o caso modelo, N 3) e um conjunto D de operadores lineares limitados em H; além do conceito de convergência D-fraca. O principal resultado desta teoria é, em certo sentido, uma generalização do célebre Teorema de Banach- Alaoglu-Bourbaki. Outra importante consequência da teoria é a equivalência entre convergência D-fraca em H1(RN); N 3; e convergência forte em Lp; para p 2 (2; 2 ) e D adequado. Com esta versão, provamos existência de solução para algumas classes de problema elípticos em domínios ilimitados, via método de minimização com vínculo.