New techniques and approaches for organic synthesis, purification and biological testing are enabling pharmaceutical industries to produce and test increasing numbers of compounds every year. Surprisingly, this has not led to more new drugs reaching the market, prompting two questions – why is there not a better correlation between their efforts and output, and can it be improved? One possible way to make the drug discovery process more efficient is to ensure, at an early stage, that the tested compounds are diverse, representative and of high quality. In addition the biological evaluation systems have to be relevant and reliable. The diversity of the tested compounds could be ensured and the reliability of the biological assays improved by using Design Of Experiments (DOE) more frequently and effectively. However, DOE currently offers insufficient options for these purposes, so there is a need for new, tailor-made DOE strategies. The aim of the work underlying this thesis was to develop and evaluate DOE approaches for diverse compound selection and efficient assay optimisation. This resulted in the publication of two new DOE strategies; D-optimal Onion Design (DOOD) and Rectangular Experimental Designs for Multi-Unit Platforms (RED-MUP), both of which are extensions to established experimental designs. D-Optimal Onion Design (DOOD) is an extension to D-optimal design. The set of possible objects that could be selected is divided into layers and D-optimal selection is applied to each layer. DOOD enables model-based, but not model-dependent, selections in discrete spaces to be made, since the selections are not only based on the D-optimality criterion, but are also biased by the experimenter’s prior knowledge and specific needs. Hence, DOOD selections provide controlled diversity. Assay development and optimisation can be a major bottleneck restricting the progress of a project. Although DOE is a recognised tool for optimising experimental systems, there has been widespread unwillingness to use it for assay optimisation, mostly because of the difficulties involved in performing experiments according to designs in 96-, 384- and 1536- well formats. The RED-MUP framework combines classical experimental designs orthogonally onto rectangular experimental platforms, which facilitates the execution of DOE on these platforms and hence provides an efficient tool for assay optimisation. In combination, these two strategies can help uncovering the crossroads between biology and chemistry in drug discovery as well as lead to higher information content in the data received from biological evaluations, providing essential information for well-grounded decisions as to the future of the project. These two strategies can also help researchers identify the best routes to take at the crossroads linking biological and chemical elements of drug discovery programs.

Quantitative structure-activity relationships (OSAR) constitute empirical analogy models connecting chemical structure and biological activity. The analogy approach to QSAR assume that the factors important in the biological system also are contained in chemical model systems. The development of a QSAR can be divided into subproblems: 1. to quantify chemical structure in terms of latent variables expressing analogy, 2. to design test series of compounds, 3. to measure biological activity and 4. to construct a mathematical model connecting chemical structure and biological activity. In this thesis it is proposed that many possibly relevant descriptors should be considered simultaneously in order to efficiently capture the unknown factors inherent in the descriptors. The importance of multivariately and multipositionally varied test series is discussed. Multivariate projection methods such as PCA and PLS are shown to be appropriate far QSAR and to closely correspond to the analogy assumption. The multivariate analogy approach is applied to a beta- adrenergic agents, b haloalkanes, c halogenated ethyl methyl ethers and d four different families of peptides. / <p>Diss. (sammanfattning) Umeå : Umeå universitet, 1986, härtill 8 uppsatser</p> / digitalisering@umu

Recently HER3, member of the epidermal growth factor receptor family (EGFR), has been found to play a crucial role in the development of resistance towards inhibitors that are given to patients with HER1- and HER2-driven cancers. As HER3 is up-regulated or over-activated in several types of human cancers, it is of outmost importance that new innovative drugs target its oncologic activity. The Affibody® Molecule Z08698 inhibits the heregulin induced signalling of HER3 with high affinity (KD~50 pM). As the Affibody® Molecule is small, has high solubility and outstanding folding kinetics, an effective penetration of tumour tissue is suggested together with a rationalized manufacturing process. Further coupling to an albumin binding domain (ABD) expands the plasma half-life of the molecule, hence increasing the molecule's potential of serving as a therapeutic. A process development for production of Z08698-VDGS-ABD094 has been established, where the molecule is efficiently produced in the E. coli host strain BL21(DE3), through a T7 based expression system. Cultivations were performed with a fed-batch fermentation process and the conditions were further optimized in order to obtain highest expression, while avoiding undesirable modifications like gluconoylations. By employing Design of experiments in combination with multivariate data analysis, a production process resulting in ~3.5 g product/ l culture could be verified. Moreover, thermolysis was evaluated as a suitable method for cell disruption, enabling an easy and cost-effective manufacturing process of the ABD fused Affibody® Molecule.

Die vorliegende Arbeit befaßt sich mit instationären, uni- oder multivariaten Zeitreihen, die bei der Beobachtung komplexer nichtlinearer dynamischer Systeme entstehen und sich der Modellierung durch ein globales Modell entziehen. In vielen natürlichen oder gesellschaftlichen Prozessen kann man jedoch wiederkehrende Phänomene beobachten, die von deren Rhythmen beeinflußt sind; ebenso lassen sich in technischen Prozessen beispielsweise aufgrund einer bedarfsorientierten Steuerung wiederholte, aber nicht periodische Verhaltensweisen ausmachen. Für solche Systeme und Zeitreihen wird deshalb vorgeschlagen, eine partielle Modellierung durch mehrere lokale Modelle vorzunehmen, die wiederkehrende Phänomene in Form zeitlich begrenzter Muster beschreiben. Um den Unwägbarkeiten dieser und sich anschließender Aufgabenstellungen Rechnung zu tragen, werden in dieser Arbeit durchgehend unscharfe Ansätze zur Modellierung von Mustern und ihrer Weiterverarbeitung gewählt und ausgearbeitet. Die Aufgabenstellung der Erkennung von Mustern in fortlaufenden Zeitreihen wird dahingehend verallgemeinert, daß unvollständige, sich noch in Entwicklung befindliche Musterinstanzen erkannt werden können. Basierend auf ebendieser frühzeitigen Erkennung kann der Verlauf der Zeitreihe -- und damit das weitere Systemverhalten -- lokal prognostiziert werden. Auf Besonderheiten und Schwierigkeiten, die sich aus der neuartigen Aufgabe der Online-Erkennung von Mustern ergeben, wird jeweils vermittels geeigneter Beispiele eingegangen, ebenso die praktische Verwendbarkeit des musterbasierten Vorhersageprinzips anhand realer Daten dokumentiert. / This dissertation focuses on non-stationary multivariate time series stemming from the observation of complex nonlinear dynamical systems. While one global model for such systems and time series may not always be feasible, we may observe recurring phenomena (patterns) in some of these time series. These phenomena might, for example, be caused by the rhythms of natural or societal processes, or a demand-oriented control of technical processes. For such systems and time series a partial modelling by means of multiple local models is being proposed. To cope with the intrinsic uncertainties of this task, fuzzy methods and models are being used throughout this work. Means are introduced for modelling and recognition of patterns in multivariate time series. Based on a novel method for the early recognition of incomplete patterns in streaming time series, a short-time prediction becomes feasible. Peculiarities and intrinsic difficulties of an online recognition of incomplete patterns are being discussed with the help of suitable examples. The usability of the pattern-based prediction approach is being demonstrated by means of real-world data.

Die Arbeit ordnet sich in das Gebiet der unscharfen Klassifikation ein und stellt im Detail eine Weiterführung der Forschung zur Fuzzy-Pattern-Klassifikation dar. Es handelt sich dabei um eine leistungsfähige systemtheoretische Methodik zur klassifikatorischen Modellierung komplexer, hochdimensionaler, technischer oder nichttechnischer Systeme auf der Basis von metrischen Messgrößen und/oder nichtmetrischen Experten-Bewertungen. Die Beschreibung der Unschärfe von Daten, Zuständen und Strukturen wird hierbei durch einen einheitlichen Typ einer Zugehörigkeitsfunktion des Potentialtyps realisiert. Ziel der Betrachtungen ist die weiterführende Nutzung des bestehenden Klassenmodells zur unscharfen Beschreibung nichtkonvexer Objektmengenmorphologien. Ausgehend vom automatischen datengetriebenen Aufbau der konvexen Klassenbeschreibung, deren vorteilhaften Eigenschaften sowie Defiziten wird im Rahmen der Arbeit eine Methodik vorgestellt, die eine Modellierung beliebiger Objektmengenmorphologien erlaubt, ohne das bestehende Klassifikationskonzept zu verlassen. Kerngedanken des Vorgehens sind: 1.) Die Aggregation von Fuzzy-Pattern-Klassen auf der Basis so genannter komplementärer Objekte. 2.) Die sequentielle Verknüpfung von Fuzzy-Pattern-Klassen und komplementären Klassen im Sinne einer unscharfen Mengendifferenz. 3.) Die Strukturierung des Verknüpfungsprozesses durch die Clusteranalyse von Komplementärobjektmengen und damit der Verwendung von Konfigurationen aus komplementären Fuzzy-Pattern-Klassen. Das dabei gewonnene nichtkonvexe Fuzzy-Klassifikationsmodell impliziert eine Vernetzung von Fuzzy-Klassifikatoren in Form von Klassifikatorbäumen. Im Ergebnis entstehen Klassifikatorstrukturen mit hoher Transparenz, die - neben der üblichen zustandsorientierten klassifikatorischen Beschreibung in den Einzelklassifikatoren - zusätzliche Informationen über den Ablauf der Klassifikationsentscheidungen erfassen. Der rechnergestützte Entwurf und die Eigenschaften der entstehenden Klassifikatorstruktur werden an akademischen Teststrukturen und realen Daten demonstriert. Die im Rahmen der Arbeit dargestellte Methodik wird in Zusammenhang mit dem Fuzzy-Pattern-Klassifikationskonzept realisiert, ist jedoch aufgrund ihrer Allgemeingültigkeit auf eine beliebige datenbasierte konvexe Klassenbeschreibung übertragbar. / This work contributes to the field of fuzzy classification. It dedicates itself to the subject of "Fuzzy-Pattern-Classification", a versatile method applied for classificatory modeling of complex, high dimensional systems based on metric and nonmetric data, i.e. sensor readings or expert statements. Uncertainties of data, their associated morphology and therewith classificatory states are incorporated in terms of fuzziness using a uniform and convex type of membership function. Based on the properties of the already existing convex Fuzzy-Pattern-Class models and their automatic, data-driven setup a method for modeling nonconvex relations without leaving the present classification concept is introduced. Key points of the elaborated approach are: 1.) The aggregation of Fuzzy-Pattern-Classes with the help of so called complementary objects. 2.) The sequential combination of Fuzzy-Pattern-Classes and complementary Fuzzy-Pattern-Classes in terms of a fuzzy set difference. 3.) A clustering based structuring of complementary Fuzzy-Pattern-Classes and therewith a structuring of the combination process. A result of this structuring process is the representation of the resulting nonconvex fuzzy classification model in terms of a classifier tree. Such a nonconvex Fuzzy-Classifier features high transparency, which allows a structured understanding of the classificatory decision in working mode. Both the automatic data-based design as well as properties of such tree-like fuzzy classifiers will be illustrated with the help of academic and real word data. Even though the proposed method is introduced for a specific type of membership function, the underlying idea may be applied to any convex membership function.

La caractérisation de texture est centrale dans de nombreuses applications liées au traitement d’images. L’analyse de textures peut être envisagée dans le cadre mathématique de l’analyse multifractale qui permet d’étudier les fluctuations de la régularité ponctuelle de l’amplitude d’une image et fournit les outils pratiques pour leur évaluation grâce aux coefficients d’ondelettes ou aux coefficients dominants. Bien que mise à profit dans de nombreuses applications, l’analyse multifractale souffre à présent de deux limitations majeures. Premièrement, l’estimation des paramètres multifractaux reste délicate, notamment pour les images de petites tailles. Deuxièmement, l’analyse multifractale a été jusqu’à présent uniquement considérée pour l’analyse univariée d’images, alors que les données à étudier sont de plus en plus multivariées. L’objectif principal de cette thèse est la mise au point de contributions pratiques permettant de pallier ces limitations. La première limitation est abordée en introduisant un modèle statistique générique pour le logarithme des coefficients dominants, paramétrisé par les paramètres multifractaux d’intérêt. Ce modèle statistique permet de contrebalancer la variabilité résultant de l’analyse d’images de petite taille et de formuler l’estimation dans un cadre bayésien. Cette approche aboutit à des procédures d’estimation robustes et efficaces, que ce soit pour des images de petites ou grandes tailles. Ensuite, l’analyse multifractale d’images multivariées est traitée en généralisant ce cadre bayésien à des modèles hiérarchiques capables de prendre en compte l’hypothèse d’une évolution lente des propriétés multifractales d’images multi-temporelles ou multi-bandes. Ceci est réalisé en définissant des lois a priori reliant les propriétés dynamiques des paramètres multifractaux des différents éléments composant le jeu de données. Différents types de lois a priori sont étudiés dans cette thèse au travers de simulations numériques conduites sur des images multifractales multivariées synthétiques. Ce travail est complété par une étude du potentiel apport de l’analyse multifractale et de la méthodologie bayésienne proposée pour la télédétection à travers l’exemple de l’imagerie hyperspectrale. / Texture characterization is a central element in many image processing applications. Texture analysis can be embedded in the mathematical framework of multifractal analysis, enabling the study of the fluctuations in regularity of image intensity and providing practical tools for their assessment, the coefficients or wavelet leaders. Although successfully applied in various contexts, multi fractal analysis suffers at present from two major limitations. First, the accurate estimation of multifractal parameters for image texture remains a challenge, notably for small sample sizes. Second, multifractal analysis has so far been limited to the analysis of a single image, while the data available in applications are increasingly multivariate. The main goal of this thesis is to develop practical contributions to overcome these limitations. The first limitation is tackled by introducing a generic statistical model for the logarithm of wavelet leaders, parametrized by multifractal parameters of interest. This statistical model enables us to counterbalance the variability induced by small sample sizes and to embed the estimation in a Bayesian framework. This yields robust and accurate estimation procedures, effective both for small and large images. The multifractal analysis of multivariate images is then addressed by generalizing this Bayesian framework to hierarchical models able to account for the assumption that multifractal properties evolve smoothly in the dataset. This is achieved via the design of suitable priors relating the dynamical properties of the multifractal parameters of the different components composing the dataset. Different priors are investigated and compared in this thesis by means of numerical simulations conducted on synthetic multivariate multifractal images. This work is further completed by the investigation of the potential benefit of multifractal analysis and the proposed Bayesian methodology for remote sensing via the example of hyperspectral imaging.

Made available in DSpace on 2016-06-02T19:50:12Z (GMT). No. of bitstreams: 1 3762.pdf: 1095394 bytes, checksum: 0ba952111fc170c93998133ca7ab7dc9 (MD5) Previous issue date: 2011-06-29 / The latest models on Knowledge Management (KM) recognize four stages to their practice: acquisition, storage, distribution and use of knowledge that presents itself as tacit and explicit. The KM is characterized as a multidisciplinary discipline, involving aspects of technical and social order and due to the diversity of issues about this topic, it is essential to delimit their study. In this sense, the proposed delimitation for this research is social and coordination, involving the identification of contextual factors, developed internally in organizations that support the process of KM in automotive companies. These contextual factors are defined based on organizational aspects to enable the development of the four phases of the KM, especially the development of human resources, organizational culture, teamwork, organizational structure and the way how knowledge is developed and absorbed internally. This thesis also identifies groupings (clusters) composed of companies with similar characteristics as the development of these contextual factors, analyzing the implications of the clusters features for KM and for the use of knowledge. The automotive companies can be characterized as advanced for improvement and innovation processes, and thus the knowledge management is fundamental to the sustainability of these activities. To achieve this goal, this research uses a quantitative research method, based on a survey. Data analysis is performed using multivariate statistical techniques of factor analysis, cluster analysis and discriminant analysis. In regarding the results were collected eight contextual factors that support the process of GC, and also identified four clusters for the development of these factors: innovate, exploited, exploited and Laggards. The companies of the first two groups have a better capacity to absorb knowledge and build a broader base of primary knowledge, making them more innovative, and the other two groups take a more reactive posture within the industry, following the development proposed by the companies of first two clusters. In the latter two groups, in particular exploited, there is the development of factors related to incremental improvement and teamwork, showing a tendency to use the knowledge designed to improve production efficiency. / Os modelos mais recentes sobre Gestão do Conhecimento (GC) reconhecem quatro fases para a sua prática: aquisição, armazenamento, distribuição e utilização do conhecimento, que se apresenta na forma tácita e explícita. A GC se caracteriza como uma disciplina multidisciplinar, envolvendo aspectos de ordem técnico e social e devido à diversidade de assuntos que cerca este tema, é essencial delimitar seu estudo. Neste sentido, o recorte proposto para esta pesquisa é social e de coordenação, envolvendo a identificação de fatores contextuais, desenvolvidos internamente nas organizações, que sustentam o processo de GC em empresas do setor automobilístico. Estes fatores contextuais são definidos a partir de aspectos organizacionais que possibilitem o desenvolvimento das quatro fases do processo de GC, destacando-se o desenvolvimento de recursos humanos, a cultura organizacional, o trabalho em equipe, a estrutura organizacional e a forma como o conhecimento é desenvolvido e absorvido internamente. Esta tese identifica também agrupamentos (clusters), compostos de empresas com características similares quanto ao desenvolvimento destes fatores contextuais, analisando as implicações das características dos clusters para a GC e também para a utilização do conhecimento. As empresas do setor automobilístico, foco desta pesquisa, podem ser caracterizadas como maduras em processos de melhoria e inovação, e, desta forma, o gerenciamento do conhecimento é fundamental para a sustentabilidade destas atividades. Para atingir este objetivo, é utilizado um método de pesquisa quantitativo, baseado em uma pesquisa survey. A análise dos dados é realizada por meio de técnicas estatísticas multivariadas de análise fatorial, análise de cluster e análise discriminante. Quanto aos resultados, foram levantados oito fatores contextuais que sustentam o processo de GC, e também identificados quatro agrupamentos de empresas quanto ao desenvolvimento destes fatores: Inovativo, Explorativo, Explotativo e Retardatário. As empresas dos dois primeiros agrupamentos possuem melhor capacidade de absorver conhecimento e de construir uma base mais ampla de conhecimento primário, tornando-as mais inovativas; e os outros dois agrupamentos assumem uma postura mais reativa dentro do setor, seguindo o desenvolvimento proposto pelas empresas dos dois primeiros clusters. Nestes dois últimos agrupamentos, em especial no explotativo, destaca-se o desenvolvimento de fatores relacionados à melhoria incremental e trabalho em equipe, denotando uma tendência de utilização do conhecimento voltada para a melhoria da eficiência produtiva.

Made available in DSpace on 2016-06-02T19:50:14Z (GMT). No. of bitstreams: 1 4067.pdf: 1964909 bytes, checksum: 20668b126678ef263e1b39495a60baa3 (MD5) Previous issue date: 2011-11-28 / Projects of capital goods with Engineering-to-Order (ETO) productive typology are usually complex due to its size and level of customization and technology. Because of these aspects the achievement of success demands a management effort and can be conditioned by several elements, among which are the Critical Success Factors (CSF). In this sense, the proposal for this research involves the identification and analysis of the relationship of these factors with the success obtained in ETO projects using a multidimensional concept that involves efficiency, organizational learning, preparing for the future and customer satisfaction. From these principles a survey was conducted involving 67 companies and 182 projects with above characteristics. Data analysis was initially performed by the technique of Factorial analysis, whose results led to combine 38 Critical Success Factors identified in the literature on 12 new factors with better ability to explain the observed phenomenon, namely: Uncertainty management, Organization and Planning, Productivity management, Robustness of the team, Management of technical performance, Learning and adaptation, Integration with external stakeholders, Reliability of the team, Schedule optimization, Precaution against technical failures, Guidance capacity and Precaution against supply failure. These new factors led to an analysis by multinomial logistic regression technique, which identified significant factors to explain each of the success dimensions. For the efficiency dimension this factors are Planning and organization, Reliability of the team, the Guidance capacity, Integration with external stakeholders, the Technical performance management and Schedule optimization, to the extent organizational learning are the optimization schedule, Productivity management and Precaution against technical failure. To the preparing for the future dimension this factors are the Management of the technical performance, the Reliability of the team, Schedule optimization, the Precaution against technical failures, the Learning and adaptation, and the Organization and Planning, the Guidance capacity. For the customer satisfaction are the Reliability of the team, the Learning and adaptation, Precaution against failure of supply, Organization and planning, Precaution against technical failures and Contingency management. Thus, from a scientific standpoint, the results obtained in this thesis extends the understanding of the topic studied and provide from the managerial point of view a framework that can guide and support decision making by managers of this type of project. / Projetos de bens de capital com tipologia produtiva Engineering-to-Order (ETO) são usualmente complexos devido a seu porte e nível de customização e tecnologia. Devido a estes aspectos a obtenção de sucesso pode ser condicionada por diversos elementos, entre os quais estão os Fatores Críticos de Sucesso (FCS). Neste sentido, o recorte proposto para esta pesquisa envolve a identificação e a análise das relações destes fatores com o sucesso de projetos ETO de forma multidimensional, envolvendo a eficiência, o aprendizado organizacional, a preparação para o futuro e a satisfação dos clientes. Para tanto, foi conduzido um survey que envolveu 67 empresas e 182 projetos com as características supracitadas. A análise dos dados foi realizada inicialmente por meio da técnica de Análise fatorial, cujos resultados permitiram conjugar 38 Fatores Críticos de Sucesso identificados na literatura, em 12 novos fatores com melhor capacidade de explicar o fenômeno observado, sendo eles: Gestão de incerteza, Organização e Planejamento, Gestão da produtividade, Robustez da equipe, Gestão de desempenho técnico, Aprendizado e adaptação, Integração com Stakeholders externos, Confiabilidade da equipe, Otimização de cronograma, Precaução contra falhas técnicas, Capacidade de orientação e Precaução contra falhas de fornecimento. A partir destes novos fatores conduziu-se uma análise por meio da técnica de Regressão logística multinomial, que permitiu identificar fatores significativos para explicar cada uma das dimensões de sucesso. Assim, para a dimensão eficiência estes fatores são a Organização e planejamento, a Confiabilidade da equipe, a Capacidade de orientação, a Integração com stakeholders externos, a Gestão de desempenho técnico e a Otimização de cronograma, para a dimensão aprendizado organizacional são a Otimização de cronograma, a Gestão de desempenho técnico e a Precaução contra falhas técnicas, para a dimensão preparação para o futuro são a Gestão de desempenho técnico, a Confiabilidade da equipe, a Otimização de cronograma, a Precaução contra falhas técnicas, o Aprendizado e adaptação, a Organização e planejamento, a Capacidade de orientação e para a dimensão satisfação do cliente são a Confiabilidade da equipe, o Aprendizado e adaptação, a Precaução contra falhas de fornecimento, a Organização e planejamento, a Precaução contra falhas técnicas e a Gestão de contingências. Assim, do ponto de vista científico, os resultados obtidos nesta tese ampliam a compreensão sobre o tema estudado e oferecem do ponto de vista gerencial um framework que pode orientar ou apoiar a tomada de decisão por parte de gerentes deste tipo de projeto.

Made available in DSpace on 2016-06-02T19:51:53Z (GMT). No. of bitstreams: 1 4153.pdf: 2182125 bytes, checksum: a074f91bda22478d2c98528b9d3de6ff (MD5) Previous issue date: 2012-02-27 / Financiadora de Estudos e Projetos / The organizations face problematic situations that involve the analysis of a significant number of correlated variables to reduce the variability of production processes. There are many approaches, methods and techniques to support problem solving, however has achieved distinction in literature and adopted by Six Sigma companies, one of whose goals is to identify and eliminate the causes of variation in product and process through the use of techniques statistics. The Multivariate Data Analysis (MDAs) belongs to a set of techniques that examine both the relationship between several variables, have not been included in intensive training programs in Six Sigma. This research is qualitative and descriptive, whose purpose is to confirm the dimensions criticism from Firka (2011) and Montgomery (2010) agree that the use of statistical methods to problems of manufacturing in the context of Six Sigma programs, through the study of multiple cases. These research findings confirm the association between the use of technical barriers in the MDAs to factors of management and sociology, such as the lack of management support, focus on short-term results, the methodological factors (selection and validation of variables and results) and the statistical assumptions (multivariate normality, multicollinearity and homoscedasticity). / As organizações enfrentam situações problemáticas que envolvem a análise de um conjunto significativo de variáveis correlacionadas para reduzir a variabilidade dos processos produtivos. Há muitas abordagens, métodos e técnicas de apoio à solução de problemas; entretanto tem obtido destaque na literatura e adotado pelas empresas o Seis Sigma, cujo um dos objetivos é identificar e eliminar as causas de variação do produto e processo por meio do uso de técnicas estatísticas. A Análise de Dados Multivariados (MDAs) pertencem a um conjunto de técnicas que analisam simultaneamente a relação existente entre diversas variáveis, ainda pouco difundidas nos programas de treinamento em Seis Sigma. Esta pesquisa tem caráter qualitativo-descritivo, cujo objetivo é confirmar as dimensões críticas apontadas por Firka (2011) e Montgomery (2010) que comprometem o uso de métodos estatísticos em problemas da manufatura no contexto dos programas Seis Sigma, por meio do estudo de casos. Os resultados desta pesquisa confirmam a associação entre as barreiras no uso de técnicas MDAs a fatores de ordem gerencial e sociológica, tais como, a falta de suporte gerencial, foco em resultados de curto prazo, fatores de ordem metodológica (seleção e validação das variáveis e dos resultados) e aos pressupostos estatísticos (Normalidade Multivariada, Multicolinearidade e Homocedasticidade). Palavras chaves: Analise de Dados Multivariados (MDAs). Fatores Críticos. Melhoria Contínua (MC). Projetos Seis Sigma.

This paper was aimed to asses and expounds the physical, chemical and microbiologic parameters which interfere in the quality of water of Sergipe River watershed. It was done from the information taken out from a research report done by FAPITEC/SE/MCT/CNPq/2007; to asses metals and trace-elements in the watershed of Japaratuba River, Sergipe River, Vaza-Barris River, Piaui River and Real River in Sergipe State, it was detailed by Researcher Doctor Professor Maria Nogueira Marques and the team of Instituto de Pesquisa Tecnologico de Sergipe – ITPS; to verify the quality of the water in those watershed, based on Water Quality Index (WQI). It was done an analysis of the data collected in this research. It was also done a multivariate data analyzed as well as main components – PCA and analysis of hierarchic group – HCA, the collected sample was analyzed based on physical, chemical and microbiologic parameters of DBO, turbidity, apparent color, pH, OD, chlorophyll-a, total dissolvent solid, oil and waxes, total phosphor, ammoniac nitrogen, CI, NO2, NO3, SO4-2, PO4-3, NA+, K+, Mg2+, Ca2+, CI, SO42-Cu, Mn, Ph, Cr, Ni, Cd, Pb, Zn, COT and thermotolerant coliform. The samples were collected in six parts of sample Sergipe River watershed in the period of dry and rainy and the samples were analyzed by the standard methods to analyze water and effluents the ―Standard Methods for the Examination of Water and Wastewater‖ (2005) – SMEWW and U.S. Environmental Protection Agency – Method 300.0 – US-EPA 300. The results present high value of concentration for turbidity, apparent color, pH, dissolved oxygen, conductibility, total hardness, oil and waxes, total alkalinity and total dissolved solid; great ions: Na+, K+, Mg2+, Ca2+, CI, SO42-, chlorophyll-a and nutrients; total phosphor, total nitrogen: N-NH3, NO3-, NO2-, PO43-, TOC, DBO, DQO, coliform and Ph, all of them over the maximum value allowed by the resolution nº 357/2005 from Conselho Nacional de Meio Ambiente – CONAMA for rivers type 1 and 2, which classify the water destined for human consume. With the results collected after adding of analysis PCA and HCA, it was identified that tendency of possible groups with point P1 and P5 and the parts P3 and P4 for every parameters analyzed altogether. Even analyzing the parameters according to matrix of correlation, it is possible to observe tha the same points self-separated indicating in the results the contamination which occur in the area of Sergipe River watershed, for anthropogenic actions due to the use and occupation of the ground, indicating the necessity of developing actions to control and tentative of shorten the impact generated because of those actors of pollution. / Este trabalho teve como objetivo avaliar e interpretar os parâmetros físicos, químicos e microbiológicos que interferem na qualidade das águas da bacia Hidrográfica do Rio Sergipe. Foi elaborado com informações extraídas do relatório de pesquisa contratado pela FAPITEC/SE /MCT/CNPq/2007; para a avaliação de metais e elementos-traço em águas das bacias hidrográficas dos rios Japaratuba, Sergipe, Vaza-Barris, Piauí e Real no Estado de Sergipe, elaborado pela pesquisadora prof. ª Dr.ª Maria Nogueira Marques e equipe do Instituto de Pesquisa Tecnológico de Sergipe – ITPS; para verificar a qualidade da água nas referidas bacias hidrográficas, baseados nos conceitos de Índices de Qualidade da água (IQA). Foi aplicado aos dados analisados das coletas realizadas nesta pesquisa, a análise multivariada de dados, como a análise de componentes principais - PCA e análise de agrupamento hierárquica - HCA, as amostras coletadas, foram analisadas quanto aos parâmetros físicos, químicos e microbiológicos de DBO, turbidez, cor aparente, pH, OD, clorofila-a, sólidos dissolvidos totais, óleos e graxas, fósforo total, nitrogênio amoniacal, Cl-, NO2-, NO3-, SO42-, PO43-, Na+, K+, Mg 2+,Ca 2+,Cl-,SO42-Cu, Mn, Fe, Cr, Ni, Cd, Pb, Zn, COT e coliformes termotolerantes. As amostras foram coletadas em seis pontos de amostragem da bacia hidrográfica do rio Sergipe, no período seco e chuvoso, e analisadas utilizando os métodos de padrões para análise de águas e efluentes o ―Standard Methods for the Examination of Water and Wastewater‖ (2005) -SMEWW e U.S. Environmental Protection Agency- Method 300.0 - US-EPA 300. Os resultados mostram altos valores de concentração para turbidez, cor aparente, pH, oxigênio dissolvido, condutividade, dureza total, óleos e graxas, alcalinidade total e sólidos dissolvidos totais; íons maiores: Na+, K+, Mg 2+,Ca 2+,Cl-,SO42-; clorofila-a e nutrientes: fósforo total, nitrogênio total, N - NH3, NO3-, NO2- , PO43-; TOC, DBO, DQO, Coliformes e Fe, todos muito acima dos valores máximos permitidos pela resolução nº: 357/2005 do Conselho Nacional de Meio Ambiente - CONAMA para rios de classe 1 e 2, que classifica as águas destinadas ao consumo humano. Com os resultados obtidos após a aplicação das análises PCA e HCA, identificou- se que a tendência dos possíveis grupos com os pontos P1 e P5 e os pontos P3 e P4 para todos os parâmetros analisados juntos. Mesmo analisando os parâmetros de acordo com as matrizes de correlação, pode-se observar que os mesmos pontos se separam indicando nos resultados a contaminação que ocorre na área da bacia hidrográfica do rio Sergipe, por ações antropogênicas devido ao uso e ocupação do solo, indicando a necessidade de desenvolver ações para o controle e tentativa de minimização dos impactos gerados por estes atores de poluição.