Model discrimination deals with situations where there are several candidate models available to represent a system. The objective is to find the “best” model among rival models with respect to prediction of system behavior. Empirical and mechanistic models are two important categories of models. Mechanistic models are developed based on physical mechanisms. These types of models can be applied for prediction purposes, but they are also developed to gain improved understanding of the underlying physical mechanism or to estimate physico-chemical parameters of interest. When model discrimination is applied to mechanistic models, the main goal is typically to determine the “correct” underlying physical mechanism. This study focuses on mechanistic models and presents a model discrimination procedure which is applicable to mechanistic models for the purpose of studying the underlying physical mechanism. Obtaining the data needed from the real system is one of the challenges particularly in applications where experiments are expensive or time consuming. Therefore, it is beneficial to get the maximum information possible from the real system using the least possible number of experiments. In this research a new approach to model discrimination is presented that takes advantage of Monte Carlo (MC) methods. It combines a design of experiments (DOE) method with an adaptation of MC model selection methods to obtain a sequential Bayesian Markov Chain Monte Carlo model discrimination framework which is general and usable for a wide range of model discrimination problems. The procedure has been applied to chemical engineering case studies and the promising results have been discussed. Four case studies, order of reaction, rate of FeIII formation, copolymerization, and RAFT polymerization, are presented in this study. The first three benchmark problems allowed us to refine the proposed approach. Moreover, applying the Sequential Bayesian Monte Carlo model discrimination framework in the RAFT problem made a contribution to the polymer community by recommending analysis an approach to selecting the correct mechanism.

In this thesis, it is aimed to design lithium extraction from boron clays using statistical design of experiments and robust design methodologies. There are several factors affecting extraction of lithium from clays. The most important of these factors have been limited to a number of six which have been gypsum to clay ratio, roasting temperature, roasting time, leaching solid to liquid ratio, leaching time and limestone to clay ratio. For every factor, three levels have been chosen and an experiment has been designed. After performing three replications for each of the experimental run, signal to noise ratio transformation, ANOVA, regression analysis and response surface methodology have been applied on the results of the experiments. Optimization and confirmation experiments have been made sequentially to find factor settings that maximize lithium extraction with minimal variation. The mean of the maximum extraction has been observed as 83.81% with a standard deviation of 4.89 and the 95% prediction interval for the mean extraction is (73.729, 94.730). This result is in agreement with the studies that have been made in the literature. However / this study is unique in the sense that lithium is extracted from boron clays by using limestone directly from the nature, and gypsum as a waste product of boric acid production. Since these two materials add about 20% cost to the extraction process, the results of this study become important. Moreover, in this study it has been shown that statistical design of experiments help mining industry to reduce the need for standardization.

Selective laser sintering (SLS) is a rapid prototyping method in which three-dimensional objects are constructed by sintering thin layers of a variety of powdered materials via laser beam. In SLS, as in most other Rapid Prototyping methods, the produced parts exhibit varying degrees of intrinsic porosity due to the discrete nature of layer-by-layer production. Selective scanning and discrete bonding of individual particles or clusters of particles impart local porosity, which is mostly an undesired trait as the part integrity decreases with increased porosity. However, there are a number of emerging or potential applications as in tissue engineering and composite/functionally graded materials, in which part porosity and its control during production are needed. In this study, the manufacturing capabilities of selective laser sintering are investigated towards producing predesigned porous structures using a polymeric powder. The porous structures are characterized in two main categories: regular porous structures, which involve geometries such as predesigned holes and lattice structures that have orderly porous architecture, and irregular porous structures, which exhibit random pore architecture that is intrinsic in all SLS parts. The limitations of producing regular porous structures are investigated, identified and quantified, based on hole size and dimensional accuracy. An experimental analysis based on design of experiments is employed to investigate the effects of processing parameters on the resulting macroscopic pore properties of irregular porous structures. A mathematical relation is developed to quantify and predict the relations between the SLS process parameters: Laser power, hatching distance, laser scan spacing, and the resulting apparent mass density (as a measure of porosity). The subsequent tests verify accuracy of the developed empirical model.

The U.S. Food and Drug Administration has accepted the guidelines put forth by the International Conference on Harmonization (ICH-Q8) that allow for operational flexibility within a validated design space. These Quality by Design initiatives have allowed drug manufacturers to incorporate more rigorous scientific controls into their production streams. <br>Fully automated control systems can incorporate information about a process back into the system to adjust process variables to consistently hit product quality targets (feedback control), or monitor variability in raw materials or intermediate products to adjust downstream manufacturing operations (feedforward control). These controls enable increased process understanding, continuous process and product improvement, assurance of product quality, and the possibility of real-time release. Control systems require significant planning and an initial investment, but the improved product quality and manufacturing efficiency provide ample incentive for the expense. <br>The fluid bed granulation and drying unit operation was an excellent case study for control systems implementation because it is a complex unit operation with dynamic powder movement, high energy input, solid-liquid-gas interactions, and difficulty with scale-up development. Traditionally, fluid bed control systems have either used first principle calculations to control the internal process environment or purely empirical methods that incorporate online process measurements with process models. This dissertation was predicated on the development of a novel hybrid control system that combines the two traditional approaches. <br>The hybrid controls reduced the number of input factors for the creation of efficient experimental designs, reduced the variability between batches, enabled control of the drying process for a sensitive active pharmaceutical ingredient, rendered preconditioned air systems unnecessary, and facilitated the collection of data for the development of process models and the rigorous calculation of design spaces. Significant variably in the inlet airstream was able to be mitigated using feedforward controls, while process analytical technology provided immediate feedback about the process for strict control of process inputs. Tolerance surfaces provided the ideal tool for determining design spaces that assured the reduction of manufacturing risk among all future batches, and the information gained using small scale experimentation was leveraged to provide efficient scale-up, making these control systems feasible for consistent use. / Mylan School of Pharmacy and the Graduate School of Pharmaceutical Sciences / Pharmaceutics / PhD / Dissertation

Nesta tese desenvolveram-se projetos e modelos de indutores planares para circuitos integrados de radiofrequência nas arquiteturas espiral e cross para as tecnologias CMOS AMS 0,35 ?m e IBM 0,13 ?m, visando representar o desempenho destes na faixa de frequências de corrente contínua até 90% da frequência de ressonância. A modelagem empregada utiliza um circuito elétrico equivalente com nove elementos para representar indutores, para o qual é apresentado um processo para extração dos valores dos seus componentes a partir de dados simulados ou medidos. Foi apresentada uma versão estendida da arquitetura cross original, mais flexível e tratável por algoritmos, com uma indutância 25% maior que a arquitetura espiral de mesma área. Utilizou-se planejamento de experimentos para elaborar conjuntos de amostras de indutores com dimensões geométricas variadas nas arquiteturas e tecnologias estudadas, para os quais desenvolveu-se um conjunto de algoritmos específicos para o projeto dos indutores a partir de suas dimensões. A simulação eletromagnética do layout de cada amostra gerou os resultados empregados no desenvolvimento do processo de extração dos valores dos componentes do circuito equivalente, os quais foram relacionados às dimensões geométricas dos indutores através de equações de projeto elaboradas por meio de análise de regressão multivariada e modelos lineares generalizados. Tanto o processo de extração dos valores dos componentes, como as equações de projeto foram validados estatisticamente pela comparação dos resultados das simulações dos respectivos circuitos equivalentes com os resultados das simulações eletromagnéticas dos layouts dos indutores em cada arquitetura e tecnologia, demonstrando a correspondência entre estes. Discrepâncias observadas em alta frequência entre os resultados de simulação eletromagnética dos indutores e seus respectivos circuitos equivalentes são associados a limitações do modelo de circuito usado, que não considera os fenômenos de correntes de Foucault (correntes de turbilhonamento) e efeito pelicular. O processo de modelagem desenvolvido, devido a sua generalidade, pode ser aplicado ao desenvolvimento de outros modelos de circuito elétrico para indutores, ou mesmo para modelagem de outros componentes passivos para circuitos integrados, como capacitores, resistores e transformadores. / In this thesis were developed designs and models for planar inductors for radiofrequency integrated circuits in the spiral and cross architectures for AMS 0.35 ?m and IBM 0.13 ?m CMOS technologies, aiming to represent the inductors performance in the frequency range of direct current up to 90% of resonant frequency. The employed modeling used an equivalent circuit for inductors with nine elements, for which a process for extracting the values of its components from simulated or measured data is presented. An extended version of the original cross architecture was presented, more flexible and treatable by algorithms, with an inductance 25% larger than the spiral architecture of the same area. Design of experiments was used to elaborate sets of inductor samples with different geometric dimensions in the studied architectures and technologies. Algorithms were developed to design the layout of each inductor considering its architecture and specific dimensions. The layout electromagnetic simulation of each sample generated the results used in the process developed for extracting the values of the components of the equivalent circuit model, which were related to the geometric dimensions of the inductors through design equations elaborated by multivariate regression analysis and generalized linear models. Both the component extraction process and the design equations were statistically validated by comparing the simulation results of the respective equivalent circuit models with the results of the electromagnetic simulations of the inductor layouts in each architecture and technology, demonstrating matching between them. Observed discrepancies at high frequency between the electromagnetic simulation results for the inductors and their respective equivalent circuits are associated with limitations of the circuit model used, which does not consider eddy currents and skin effect phenomena. The modeling process developed, due to its generality, can be applied to the development of other electric circuit models for inductors, or even to model other passive components for integrated circuits, such as capacitors, resistors and transformers.

With ever-increasing numbers of engine actuators to calibrate within increasingly stringent emissions legislation, the engine mapping and calibration task of identifying optimal actuator settings is much more difficult. The aim of this research is to evaluate the feasibility and effectiveness of the Multidisciplinary Design Optimisation (MDO) frameworks to optimise the multi-attribute steady state engine calibration optimisation problems. Accordingly, this research is concentrated on two aspects of the steady state engine calibration optimisation: 1) development of a sequential Design of Experiment (DoE) strategy to enhance the steady state engine mapping process, and 2) application of different MDO architectures to optimally calibrate the complex engine applications. The validation of this research is based on two case studies, the mapping and calibration optimisation of a JLR AJ133 Jaguar GDI engine; and calibration optimisation of an EU6 Jaguar passenger car diesel engine. These case studies illustrated that: -The proposed sequential DoE strategy offers a coherent framework for the engine mapping process including Screening, Model Building, and Model Validation sequences. Applying the DoE strategy for the GDI engine case study, the number of required engine test points was reduced by 30 – 50 %. - The MDO optimisation frameworks offer an effective approach for the steady state engine calibration, delivering a considerable fuel economy benefits. For instance, the MDO/ATC calibration solution reduced the fuel consumption over NEDC drive cycle for the GDI engine case study (i.e. with single injection strategy) by 7.11%, and for the diesel engine case study by 2.5%, compared to the benchmark solutions.

Esta dissertação apresenta a aplicação de um método para otimização multivariada de um processo químico, no qual são utilizadas técnicas de projetos de experimentos, modelos de regressão e função de perda multivariada para a definição do ajuste ótimo do processo de clorometilação a um custo de pesquisa mínimo. O estudo foi realizado com duas matérias-primas diferentes, DHS e DHS-S, pois atualmente é a partir delas que o produto Butóxido de Piperonila é obtido, e a origem dessas matérias-primas, a pureza e seus preços são diferenciados. O interesse da pesquisa está em encontrar um ajuste ótimo para os fatores controláveis quantidade de produtos no reator, proporção de DHS ou DHS-S, temperatura de reação e tempo de reação com vistas a melhorar a produtividade, a qualidade e reduzir o custo de produção. O trabalho mostra de forma estruturada as diversas etapas envolvidas na aplicação prática de uma otimização multivariada. Um projeto fatorial completo foi escolhido para realização dos experimentos, e os dados foram tratados estatisticamente através de modelos de regressão.Os objetivos específicos da pesquisa foram alcançados, possibilitando uma melhoria na qualidade do produto clorometilado, um aumento na produtividade do processo para a matéria-prima DHS-S em 18,25% e uma redução no custo de produção para a matéria-prima DHS de 1,11% e para a matéria-prima DHS-S de 5,11% / This paper presents the application of a method for multivariate optimization of a chemical process, emphasizing techniques of design of experiments, and models of regression and quadratic loss function for identifying excellent adjustment of process, with a minimum research cost considering multiple quality characteristics simultaneously. The study was realized considering two different raw materials, DHS and DHS-S, because nowadays Piperonyl Butoxide is obtained from them, and their purity and prices are different. The interest of the research is to find the optimum adjustment for the controllable factors quantity of products on reactor, DHS or DHS-S proportion, reaction temperature and reaction time to improve the productivity, the quality and to reduce the production costs. This paper shows, in a structured manner, several steps involved on the practice application of the multivariate optimization. A full factorial project was chosen for the experiments, and the data were treated statistically by models of regression. The specific objectives of the research were attained, permitting an improvement in the Chloromethyl quality, an increase in the productivity of the process for the DHS-S raw material in 18.25% and a reduction in the production cost for the DHS raw material of 1.11% and for the DHS-S raw material of 5.11%.

No Brasil, tanto o agronegócio quanto a fabricação de celulose trazem resultados relevantes para a economia. Sob uma perspectiva de sustentabilidade, as indústrias beneficiadoras de materiais vegetais geram grande volume de resíduos, os quais podem servir de fonte de celulose, configurando-se numa alternativa atraente para gerar lucros e diminuir impactos ambientais negativos. O objetivo geral desta dissertação é realizar um estudo exploratório de extração de celulose a partir de resíduo do processo produtivo de empresa beneficiadora de material vegetal, sob perspectiva das dimensões ambiental e econômica da sustentabilidade. O estudo foi desdobrado em cinco etapas: i) identificação na literatura das diretrizes para os processos produtivos sustentáveis, por meio das abordagens que tratam da sustentabilidade; ii) identificação dos processos industriais de extração de celulose e suas características operacionais; iii) definição da empresa para realização do estudo, a sazonalidade e o volume disponível dos resíduos vegetais provenientes do processo de beneficiamento de vegetais; iv) adaptação do processo industrial de extração de celulose para uso com resíduos vegetais e atendimento às restrições da fábrica definida para o estudo e v) realização de testes de extração com o uso de projeto de experimentos para posterior caracterização da polpa de celulose obtida do resíduo. Neste último, foi empregado um Projeto Fatorial Composto de Segunda Ordem (PSCO) com Metodologia de Superfície Resposta (MSR) para analisar a influência dos fatores controláveis sobre as variáveis de resposta. Os fatores controláveis foram definidos de acordo com dados da literatura específica e foram analisadas as variáveis de resposta que caracterizam polpa de celulose, aplicáveis a fins sanitários, como Número Kappa, capacidade de absorção, velocidade de absorção e densidade. Resultados indicaram que limitações quanto ao uso de altas temperaturas no processo de extração é um fator relevante, para atingir maiores teores de deslignificação. Os estudos mostraram a influência do uso de aditivo de processo, antraquinona, carga alcalina e tempo de polpeamento, bem como, tratamento de hidrólise ácida no resíduo na qualidade da polpa. / Agribusiness and pulp production bring both relevant results to the Brazilian economy. In a sustainable perspective, the wastes of Industries that benefit vegetable materials may serve as pulp’s source, configuring itself as an attractive alternative for profit generation and reduction of negative environmental impacts. The aim in this dissertation is to carry out an exploratory study of pulp extraction from the waste of productive process of a firm which processes vegetal material, under the perspective of environmental and economic dimensions of sustainability. The study was unfolded in five steps: i) identification, in literature, the guidelines for sustainable productive processes, through approaches that embrace sustainability; ii) identification of industrial processes of pulp extraction and respective operational features; iii) definition of an industrial unit to perform the study, as well as the seasonality and the availability of vegetal wastes coming from the vegetable process; iv) adaptation of the industrial process of pulp extraction for using with vegetable wastes, and attendance to constraints of the current process in the industrial unit chosen for the study; and v) the performance of extraction tests using a project of experiments for further characterization of the cellulose pulp resulting from the waste. It was employed a Factorial composite second-order design in a Surface Response Methodology (SRM) in order to analyze the influence of the controlled variables over the response variables. Variables of the extraction process were defined according with the specific literature’s data, and response variables, adequate to the cellulose pulp characterization as Kappa Number, capability, absortion speed and apparent density were analyzed. Results indicate that constraints related to the high temperature used in the equipment available for extraction is a relevant factor in the process for obtaining higher contents of delignification. Studies show the influence of using additive process, anthraquinone, load time and alkaline pulping, as well as treatment of hydrolysis in the waste of pulp quality. The remarkable difference in the quality of vegetable waste after hydrolysis treatment indicates that a pretreatment stage of such waste is interesting, since the fiber is resistant and requires high alkali and addictive charges.

Este trabalho tem o objetivo de quantificar os efeitos da manutenção periódica de motores de tratores agrícolas sobre a potência medida na sua tomada de potência - TDP, e sobre o consumo de combustível. Para isso, utiliza-se o método de planejamento de experimentos, com um fatorial completo 25 sem repetição, totalizando 32 ensaios independentes. Para o experimento, utiliza-se um trator agrícola de rodas, com motor 6 cilindros, de 6,8 litros, com sistema de combustível de injeção direta, com uma bomba injetora rotativa, turboalimentado e com intercooler. Os parâmetros de controle são escolhidos com base na experiência do autor e seus níveis são definidos experimentalmente. São eles: a) condição externa do radiador, b) condição do filtro de combustível, c) condição do filtro de ar, d) qualidade do combustível e e) condição do óleo de motor. É possível constatar que os parâmetros com maior influência na potência são a condição do filtro de ar, condição do óleo do motor e uma interação de segunda ordem entre a qualidade do combustível e o óleo do motor. Eles são responsáveis por variações na potência de, respectivamente, 2,7%, 2,4% e 1,7%, referenciados na condição ideal de teste. O consumo de combustível não apresenta variação significativa em relação a nenhum dos fatores estudados, o que leva a concluir que não existe relação entre ele e a manutenção periódica do trator nos termos propostos por este trabalho. Por fim, mostra-se que a correta manutenção, desde sua condição extrema até sua condição ideal, aumenta em até 7% a potência útil disponibilizada pelo equipamento para o trabalho do campo. / This study aims to quantify the effects of periodic maintenance of agricultural tractors’ engines on the power measured at its power takeoff - PTO, and its fuel consumption. To do this, we use the design of experiments method, 25 full factorial without replication, totaling 32 independent assays. For the experiment, we use an agricultural wheel tractor, with 6 cylinder engine, 6,8 liters, rotary fuel pump and direct injection fuel system, turbocharged with intercooler. The control parameters are chosen based on the author's experience and their levels are defined experimentally. They are: a) external condition of radiator, b) fuel filter condition, c) air filter condition, d) quality of the fuel and e) engine oil condition. It can be seen that the parameters with the greatest influence on power are the air filter condition, engine oil condition and a second-order interaction between fuel quality and engine oil. They are responsible for variations in power, respectively of, 2.7%, 2.4% and 1.7%, results based on ideal condition. Fuel consumption presents no significant variation in relation to any of the factors studied, which leads to the conclusion that there is no relation between it and the periodic maintenance of the tractor the way it is proposed by this work. Finally, it is shown that with the correct maintenance, since the extreme condition to the ideal condition, increases up to 7% the power provided by the equipment to work in the field.

Este trabalho de conclusão tem como tema a relação entre as especificações de resinas alquídicas e de tintas preparadas com estas resinas, com foco nas características de teor de sólidos e viscosidade, fazendo uso das técnicas projeto de experimentos, superfície de resposta e análise de regressão. O objetivo principal é o estudo dos limites de especificação ideais para a resina alquídica, de forma que os lotes de tinta apresentem propriedades dentro da especificação já ao final do processamento, reduzindo a incidência de lotes fabris com necessidade de ajuste. Como conseqüência, temos redução de retrabalho e lead time fabril, adequação de custo de produtos, maior qualidade intrínseca e maior confiança na sistemática de desenvolvimento, produção e controle de qualidade. Inicialmente, foi realizada uma revisão bibliográfica sobre a tecnologia de tintas e resinas alquídicas, conceitos de controle de qualidade, planejamento de experimentos, análise por superfície de resposta e análise de regressão. Na seqüência, foi conduzido o estudo de caso, realizado na empresa Killing S.A. Tintas e Adesivos, planta localizada na cidade de Novo Hamburgo. Os resultados experimentais indicaram modelos de regressão polinomial válidos para as propriedades avaliadas. Foram tomadas as propriedades teor de sólidos e viscosidade Copo Ford #2 da mistura A+B como parâmetros para análise dos limites de especificação da resina alquídica, onde se comprovou que a variabilidade atualmente permitida é excessiva. A aplicação dos modelos de regressão indicou novos limites de especificação para a resina alquídica, mais estreitos, viabilizando a obtenção de tintas com propriedades especificadas. / This master thesis presents an analysis of the relationship between specification of alkyd resins and relative coatings, regarding two characteristics: solids value and viscosity. The study was conducted applying the techniques of design of experiments, response surface and regression analysis. The main objective was to study the ideal specification limits to the alkyd resin, so that the coatings batches present properties equivalent to the specification reducing the occurrence of batch adjustments. Reduction in adjustments lead to reduced rework, lead time, and manufacturing costs. It also contributes to higher intrinsic quality and trust in the development, production and quality control system. The first step was a literature review about the coating and alkyd resin technology, control quality concepts, design of experiments, response surface and regression analysis. Then, a case study was developed in the company Killing S.A. Tintas e Adesivos, a producer located in the city of Novo Hamburgo. The experimental results denoted valid polynomial regression models to the evaluated properties.It was studied the properties solids value and viscosity Ford Cup #2 to the mixture A+B as parameters to the specification limits analysis of the alkyd resin, being proved that the actual permitted variability is excessive. The application of the regression models pointed narrower specification limits to the alkyd resin, permitting the production of coatings with specificated properties.