Recent technologies have changed the way companies acquire and use computing resources. Companies have to adapt their capabilities, which combine business processes, skills, etc., to exploit the opportunities presented by these technologies whilst avoiding adverse effects. The latter part is, however, becoming increasingly difficult due to the uncertain long-term impact recent technologies have. This thesis argues that companies are required to adapt their capabilities in a way that increases the company's resilience so that they are robust yet flexible enough to succeed under uncertain conditions. By focusing on cloud computing as one recent technology, this thesis first identifies the underlying processes of adapting capabilities to cloud computing by investigating how software vendors migrated their products into the cloud. The results allow the definition of viewpoints that influence the adaptation of capabilities to cloud computing. Furthermore, the Functional Resonance Analysis Method (FRAM) is applied to one software vendor after the migration of their product into the cloud. FRAM enables the analysis of ‘performance variabilities' that need to be dampened to increase the resilience of systems. The results show that FRAM appropriately informs steps to increase and measure resilience when migrating products into the cloud. The final part develops cFRAM which extends FRAM through the viewpoints to enable the analysis of capabilities within FRAM. The goal of cFRAM is to enable companies to (1) identify existing capabilities, (2) investigate the impact of cloud computing on them, and (3) inform steps to adapt them to cloud computing whilst dampening performance variabilities. The results of the cFRAM evaluation study are unequivocal and show cFRAM is a novel method that achieves its goal of enabling companies to adapt their capabilities to cloud computing in a way that increases the company's resilience. cFRAM can be easily adapted to other technologies like smartphones by changing the viewpoints.

Submitted by Maria Magnus (maria.carolina87@hotmail.com) on 2018-01-19T14:07:04Z No. of bitstreams: 2 MAGNUS_maria_carolina_machado.pdf: 2245090 bytes, checksum: bd1a8639ac7e9b106d8c8fd29bba6207 (MD5) carta-comprovante-ORIENTADOR.pdf: 231514 bytes, checksum: da86cd23ef2487bd14c076af966ee0c0 (MD5) / Approved for entry into archive by Ronildo Prado (bco.producao.intelectual@gmail.com) on 2018-01-30T13:16:11Z (GMT) No. of bitstreams: 2 MAGNUS_maria_carolina_machado.pdf: 2245090 bytes, checksum: bd1a8639ac7e9b106d8c8fd29bba6207 (MD5) carta-comprovante-ORIENTADOR.pdf: 231514 bytes, checksum: da86cd23ef2487bd14c076af966ee0c0 (MD5) / Approved for entry into archive by Ronildo Prado (bco.producao.intelectual@gmail.com) on 2018-01-30T13:16:20Z (GMT) No. of bitstreams: 2 MAGNUS_maria_carolina_machado.pdf: 2245090 bytes, checksum: bd1a8639ac7e9b106d8c8fd29bba6207 (MD5) carta-comprovante-ORIENTADOR.pdf: 231514 bytes, checksum: da86cd23ef2487bd14c076af966ee0c0 (MD5) / Made available in DSpace on 2018-01-30T13:22:10Z (GMT). No. of bitstreams: 2 MAGNUS_maria_carolina_machado.pdf: 2245090 bytes, checksum: bd1a8639ac7e9b106d8c8fd29bba6207 (MD5) carta-comprovante-ORIENTADOR.pdf: 231514 bytes, checksum: da86cd23ef2487bd14c076af966ee0c0 (MD5) Previous issue date: 2018-01-15 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / The present research was developed in 2 movements. In movement 1, I describe the history of Modeling, a story written from what counts the theses and dissertations defended in Brazil, from 1987 to 2016. The document history seeks an origin and shows the development of Modeling in a linear and evolutionary way, free of mishaps and discontinuities. In movement 2, the goal is to write the history of the present modeling - history monument - to show the conditions of possibilities for its emergence, the cracks that this discourse underwent in the course and the marks of the past in the present - its regularities. The theoretical-methodological contributions, which support the writing of monument history, are linked to the theorizations of the philosopher Michel Foucault. The analytical material covers theses and dissertations, annals of the National Conference on Modeling in Mathematics Education and interviews. The analysis of these materials evidenced that the emergence of the modeling discourse occurs in the midst of a crisis in the teaching of Mathematics, consisting of the following statements: “students have difficulty in learning Mathematics” and “Mathematics is distant from reality”. The scrutiny of the material also gives visibility to some discursive discontinuities: the modeling discourse when intertwining with the constructivist discourse, in the 1980s, displaces the sense of studying from models - product - to construct models – process; from the economic and political conditions experienced in Brazil, the discussion about the formation of a critical and reflective subject emerges in the 1990s, in the discursive plots of Modeling, evidencing that the discourse of Modeling is also an identity question. Afterwards, the analysis gives visibility to the modeling discourse as a mechanism for the privileging of the school content, showing that Modeling legitimizes and supports the curricular machinery and, also, is sustained and legitimized by it. / A presente pesquisa foi desenvolvida em 2 movimentos. No movimento 1, descrevo a história documento da Modelagem, uma história escrita a partir do que contam as teses e dissertações defendidas no Brasil, no período de 1987 a 2016. A história documento é uma história que busca uma origem e mostra o desenvolvimento da Modelagem de forma linear e evolutiva, livre de percalços e descontinuidades. No movimento 2, o objetivo é escrever a história do presente da Modelagem - história monumento -, mostrar as condições de possibilidades para sua emergência, as rachaduras que esse discurso sofreu no percurso e as marcas do passado no presente – as suas regularidades. Os aportes teóricos-metodológicos, que sustentam a escrita da história monumento, vinculam-se às teorizações do filósofo Michel Foucault, de pensadores dos Annales, tais como: Peter Burke e Jacques Le Goff, e historiadores da ciência, como Georges Canguilhem e Gaston Bachelard. O material analítico abrange teses e dissertações, anais da Conferência Nacional sobre Modelagem na Educação Matemática e conversas que tive com alguns pesquisadores. A análise desses materiais evidenciou que a emergência do discurso da Modelagem ocorre em meio a uma crise no ensino de Matemática, constituída pelos seguintes enunciados: “os alunos têm dificuldade na aprendizagem da Matemática” e “a Matemática é distante da realidade”. O escrutínio do material, também, dá visibilidade a algumas descontinuidades discursivas: o discurso da Modelagem ao se entrelaçar ao discurso construtivista, na década de 1980, desloca o sentido de estudar a partir de modelos – produto – para construir modelos – processo; a partir das condições econômicas e políticas vivenciadas no Brasil, emerge na década de 1990, nas tramas discursivas da Modelagem, a discussão em torno da formação de um sujeito crítico e reflexivo, evidenciando que o discurso da Modelagem é, também, uma questão de identidade. Posteriormente, a análise dá visibilidade ao discurso da Modelagem enquanto um mecanismo de privilegiamento dos conteúdos ditos escolares, mostrando que a Modelagem legitima e sustenta a maquinaria curricular e, também, é por ela sustentada e legitimada.

Le travail présenté dans ce mémoire consiste à adapter la méthode de corrélation d'images numériques (CIN) pour la gestion des discontinuités du milieu et de la transformation. Le cadre d'utilisation des méthodes optiques de mesure i.e. celui de la mécanique des milieux continus, impose une continuité matérielle : i) du domaine et ii) de la transformation. Pour traiter les discontinuités en question, on peut reconsidérer la continuité (milieu et transformation) par morceaux. Dans le cadre de la CIN, cela se traduit par la possibilité d'adapter localement la forme et la taille des fenêtres de corrélation à la zone d'intérêt et à sa cinématique. Pour ce faire, il est possible d'utiliser des masques : un masque-objet pour traiter les discontinuités du milieu et un masque-discontinuité pour traiter les discontinuités de la transformation. Cependant, avant l'implémentation des masques dans la procédure de corrélation, plusieurs essais expérimentaux de déplacement sur des modèles comportant les deux types de discontinuité, ont été conduits. Ces essais ont permis de prouver l'influence des discontinuités sur la dégradation de la précision de mesure par CIN. Ensuite, des essais de traction ont été menés sur une éprouvette en polycarbonate conduisant à un mode I d'ouverture ou de fermeture de fissure. Après l'implémentation des masques dans le logiciel de corrélation Correla, le calcul des champs de déplacement et de déformation a été opéré avec succès à la surface de cette éprouvette, démontrant ainsi l'efficacité de la CIN adaptée. / The development of the digital image correlation method (DIC) for the management of discontinuities of the material and the transformation is discussed in this thesis. As we know, the framework for the use of the optical measurement methods i.e. of continuum mechanics requires a continuity of: i) domain and ii) transformation. To treat those discontinuities, we can consider a piecewise continuity (material and transformation). In the case of DIC method, this can be done by adapting locally the shape and size of the correlation subsets to the zone of interest and its kinematics. A novel way to do it, is by using masks: an object mask to process material discontinuities and a discontinuity mask to process transformation discontinuities. However before the implementation of masks in the correlation process, several experimental displacement tests on models reproducing the two types of discontinuity at small scale were performed. Those tests proved the influence of discontinuities on the degradation of the measurement accuracy by DIC. Then, tensile tests were conducted on a polycarbonate made specimen knowing a mode I opening or closing crack. After implementing masks in the correlation software Correla, the calculation of displacement and deformation fields was successfully performed on the surface of this specimen for all its edges, demonstrating the effectiveness of the adapted DIC.

Ce mémoire porte sur la reconstruction 3D active par la lumière structurée et se veut une introduction au travers d'une vue d'ensemble du domaine. L'accent est mis sur une compréhension des différents aspects du scan 3D, en commençant par l'acquisition du matériel et en allant jusqu'au calcul d'un modèle 3D. Les défis techniques les plus fréquents dans un contexte de scan 3D avec de l'équipement à faible coût sont abordés, de façon à donner une idée générale des problèmes à anticiper à quelqu'un qui souhaiterait se lancer dans la conception ou l'utilisation de ce type de scanneur. Afin d'aider à peser les forces et faiblesses de chaque algorithme formant l'état de l'art actuel en lumière structurée, une revue des différentes stratégies existantes est également donnée. L'étude du domaine a mené à une contribution sous la forme d'un article, qui a été intégré dans le chapitre 6. Cet article propose une amélioration en ce qui concerne la performance et la précision d'un algorithme de scan 3D existant et a été publié dans la International Virtual Conference on 3D Vision en novembre 2020. / This thesis focuses on active 3D reconstruction with structured light and is intended to be an introduction through an overview of the field. It is focused on understanding the different aspects of a 3D scanner, from the acquisition of the material to the final mesh reconstruction. The most common technical difficulties that arise in the context of low-cost 3D scanning are addressed, to give a general idea of the potential technical problems that may occur when using a DIY 3D scanner built from scratch. To help assess the strengths and weaknesses of each algorithm of the current state of the art in structured light, a review of the different strategies is also proposed. The study of this domain led to the contribution of a paper, presented here in its entirety in chapter 6. This paper describes an improvement of an existing algorithm, both in terms of performance and precision, and was published in the International Virtual Conference on 3D Vision in November 2020.

This dissertation is concerned with invertibility and one-sided invertibility of Toeplitz operators over the half plane whose generating functions admit homogenous discontinuities, and with stability of their pseudo finite sections. The invertibility criterium is given in terms of invertibility of a family of one dimensional Toeplitz operators with piecewise continuous generating functions. The one-sided invertibility criterium is given it terms of constraints on the partial indices of certain Toeplitz operator valued function.

Our sample (Xit; Yit) consists of pairs of variables. The real variable Xit measures the creditworthiness of individual i in period t. The Bernoulli variable Yit is the default indicator of individual i in period t. The objective is to estimate a credit rating system, i.e. to particularly divide the range of the creditworthiness into several rating classes, each with a homogeneous default risk. The field of change point analysis provides a way to estimate the breakpoints between the rating classes. As yet, the literature only considers models without dependencies or with dependence only in cross section. This contribution proposes multi-period models including dependencies in cross section as well as in longitudinal section. Furthermore, estimators for the model parameters are suggested. The estimators are applied to a data set of a German credit bureau.

Detta examensarbete är en ansats mot en förenklad metod för sprickkartering i borrkärna och utvärderar huruvida en förenklad metod är tillämpbar vid projekt som anläggs under aktiv design och tid är en kritisk faktor. Det föreliggande arbetet är en studie baserad på data från Boliden Mineral AB, från undersökningar utförda i syfte att lokalisera en lämplig, lågpermeabel bergvolym för anläggandet av ett slutförvar för miljöfarligt avfall. Vattenflöde i kristallin bergmassa, liksom den i Sverige, förmodas traditionellt påverkas mest av sprickornas karaktär så som sprickdensitet, sprickvidd, råhet, sprickfyllning etc. Vidare påverkar sprickegenskaperna bergmassans hållfasthet, varför karaktärisering av bergmassan utgör en central roll vid byggande i berg. Effektivisering av sprickkartering under aktiv design undersöktes genom att utvärdera hur väl de parametrar som används vid karaktärisering verkar för identifiering av vattenförande strukturer. Vidare utvärderades möjligheten att minska antalet sprickor att kartera och karaktärisera vid den fysiska kärnan. Resultat i denna studie visar inte några tydliga kopplingar mellan mätsektionernas hydrauliska konduktivitet (K) och de individuella karterade parametrarna. Osäkerheterna är dock stora då mätsektionerna för de hydrauliska testerna utgörs av 10 m för uppskattande av K, som dessutom påverkas av storskaliga företeelser (t.ex. grad av sammankoppling) som vidare inte är möjligt att uppskatta från borrhålsdata. Fjärrkartering av borrhålsfilm underestimerar överlag bergkvaliteten för de tre borrhål som analyserades. En förenklad metod med ett representativt urval sprickor för fysisk kartering beskriver fördelningen av parametervärden väl, vid jämförelse mot en fullständig fysisk kartering för två borrkärnor. Lerfyllda sprickor och höga Ja-värden överestimeras dock vid den förenklade karteringen vilket potentiellt kan leda till en mer konservativ förstärkningsåtgärd än nödvändigt. Fler studier behöver dock genomföras för att utvärdera huruvida den förenklade metoden kan reflektera parametervärdenas fördelning för individuella sprickgrupper. Vidare krävs ytterligare undersökningar för utvärdering av hur robusta de presenterade metoderna är vid kartering av borrkärnor med högre sprickighet än de som använts för analys i denna studie. / This master thesis is the inception of a simplified method for logging fractures in drill holes and comprises an evaluation on whether a simplified method is applicable for projects where the observational method is used and time is a critical factor. The following work is a study based on data from Boliden Mineral AB, acquired from investigations for locating a suitable, low-permeable rock mass for a repository for hazardous waste. Water flow in crystalline bedrock, such as that found in Sweden, is traditionally thought to be most influenced by fracture characteristics like fracture density, aperture, roughness, fillings etc. Moreover, these same characteristics influence the rock mass strength, which is why the characterization of fractures is fundamental for construction in rock. To improve efficiency of fracture logging in projects maintained under the observational method, an evaluation on how well the quantified fracture characteristics work for identification of large-scale water bearing structures was performed. Furthermore, the possibility of decreasing the number of fractures to log and characterize in the core was investigated.  The results show no apparent correlations between hydraulic conductivity (K) in test sections and the individually mapped parameters. However, the uncertainties are significant, mainly because of the large sections in which the hydraulic tests were conducted, but also because of largescale features influencing K (e.g. interconnectivity) which are not possible to estimate from core data.  Remote logging of borehole images generally underestimates the quality of the rock mass for the three analyzed boreholes. A simplified method with a representative selection of fractures for physical core logging was found to describe the distribution of parameter values well, when compared to a complete physical core log for two drill cores. Clayfillings and high Ja -values are however overestimated when the simplified method is applied which may lead to a more conservative reinforcement measure than necessary. However, more studies need to be carried out to evaluate whether the simplified method can reflect the distribution of parameter values for individual fracture sets. Moreover, additional studies are required to evaluate the robustness of the suggested simplified methods in drill cores from areas of more heavily fractured bedrock than those analyzed in this study.

The distinct element method (DEM) is a useful tool in rock engineering to model jointed rock masses. To simulate a jointed rock mass realistically, the main challenge is to be able to capture its complex geometry which consists of blocks with various shapes and sizes, and to model the interactions between these blocks. The main contribution of this thesis is the development of novel algorithms in the DEM to model jointed rock masses, namely rock slicing procedures for block generation, and algorithms for contact detection between polygonal blocks in 2-D or polyhedral blocks in 3-D. These algorithms make use of convex optimisation techniques, for which there exist efficient solution procedures. They do not rely on conventional vertex-edge-face hierarchical data structures and tedious housekeeping algorithms. The algorithms have been verified against analytical and numerical solutions, as well as validated against experimental results published in the literature. Among those, the results of DEM simulations were compared against the experimental model tests and numerical simulations of jointed beams carried out by Talesnick et al. (2007) and Tsesarsky & Talesnick (2007) respectively. Emphasis was placed on modelling the stiffness of the block interfaces accurately, and this was accomplished by reinterpreting the laboratory data published by the investigators. The capabilities of the numerical tools are also examined and demonstrated in areas for which the DEM has found practical application. A substantial fraction of this thesis is devoted to illustrating how these tools can assist the engineer in designing support systems; for example, designing the length and spacing of rock bolts and the lining thickness for a tunnel. Algorithms to model rock bolt and lining support were implemented for this purpose. Interesting comparisons with elastic solutions for supported openings were obtained. Further, it is shown that the relative benefit of introducing more rock bolts or thicker lining can be evaluated using the numerical tools with the aid of an interaction diagram. In the final part of this thesis, the case history of the 1963 Vaiont rock slide in Italy is studied. The 2-D analyses led to useful insights concerning the influence of the reservoir water level, the rock mass strength and deformability, and the slide surface shear stiffness. 3-D analyses were undertaken to investigate the influence of the eastern boundary of the slope, and interesting insights were obtained concerning the slope kinematics. Overall, the case study shows that the tools are capable of modelling problems with specific physical and geometrical detail in both 2-D and 3-D.

André MEYNIER, héritier de la géographie rurale française, étudia, avec le concours de la télédétection, les formes des parcellaires bocagers et y détecta ce qu’il nomma des « ellipses bocagères ». Il y consacra plusieurs articles, parmi lesquels «les ensembles cadastraux circulaires en Bretagne », dans les Annales de Bretagne, (Rennes, 1945) et «la genèse du parcellaire breton » (Norois, Poitiers, 1966). Après la révolution épistémologique de la géographie française des années 1960 et 1970, cet objet d’étude tomba progressivement en désuétude, sort qui fut d’ailleurs celui de la géographie rurale morphologique dans son ensemble. Dans cette thèse, cet objet – élargi à la notion de forme spatiale héritée, donc liée au temps historique long au sens braudélien – est exhumé pour en réexaminer la pertinence au regard des paradigmes de la géographie contemporaine. Comment insérer le temps long – parfois perçu comme permanence et immobilité, du moins à l’échelle de l’actualité – dans une géographie à vocation dynamique ? Comment les nouveaux courants géographiques auraient-ils pu s’emparer de cet objet qu’un regard rapide condamne à une accusation de passéisme ? Les réflexions engagées se veulent ouvertes à tous les courants de la géographie et n’en rejettent aucun. L’objectif est aussi de développer une dialectique disciplinaire qui transgresse les paradoxes apparents. Pour explorer ces problématiques, les formes spatiales héritées sont choisies non plus en Bretagne mais en France du Nord et au Benelux, où les espaces boisés, notamment ardennais, jouent un rôle majeur dans les limites de finages ; ce sont aussi des espaces frontaliers, ce qui permet la mise en relation avec des discontinuités. Cet espace, situé au seuil du monde Hercynien vaudra quelques escapades vers l’est où il se prolonge, engageant à un voyage vers le monde slave / André MEYNIER, one of the heirs of French rural geography studied the forms in partitioned bocage landscapes and with the help of remote sensing distinguished what he called enclosed landscapes ellipses. He dedicated several articles to them among which “les ensembles cadastraux circulaires en Bretagne” in les Annales de Bretagne (Published in 1945 in Rennes) and “la genèse du parcellaire Breton” (Published in 1966 in Norois). After the epistemological revolution of French geography in the 1960’s and 1970’s, this subject gradually became obsolete as well as classical rural geography as a whole. In this thesis, the subject of my study – broadened to the notion of Inherited Spatial Forms, and consequently connected to long historical time as Fernand Braudel defines it – is reexamined to evaluate its relevance when confronted to the paradigms of modern geography. Along with a geography intending to be dynamic, how could we bring long historical time into the equation when such a time is - on a recent scale at least - sometimes perceived as still and permanent ? How would new geographical streams have been able to express interest in this subject when it is often overlooked as a remnant of the past? The present reflection is meant to encompass all geographical streams without leaving any aside. The objective is also to develop a disciplinary dialectic that would go further than visible paradoxes. In order to delve into these problems, Inherited spatial forms were chosen not only in Brittany but in Northern France and in the Benelux as well, where wooded areas on the outskirts of villages, especially in the Ardennes, have a major role on defining the limits of said villages: they are also bordering areas allowing the link with discontinuities to be made. This region on the threshold of the Hercynian world will take us to an adventurous trip through Eastern Europe

Dans cette thèse, nous étudions un principe général de convexification permettant de traiter certainsproblèmes variationnels non convexes sur Rd. Grâce à ce principe nous pouvons mettre en oeuvre lespuissantes techniques de dualité et ramener de tels problèmes à des formulations de type primal–dualdans Rd+1, rendant ainsi efficace la recherche numérique de minima globaux. Une théorie de ladualité et des champs de calibration est reformulée dans le cas de fonctionnelles à croissance linéaire.Sous certaines hypothèses, cela nous permet de généraliser un principe d’exclusion découvert parVisintin dans les années 1990 et de réduire le problème initial à la minimisation d’une fonctionnelleconvexe sur Rd. Ce résultat s’applique notamment à une classe de problèmes à frontière libre oumulti-phasique donnant lieu à des tests numériques très convaincants au vu de la qualité des interfacesobtenues. Ensuite nous appliquons la théorie des calibrations à un problème classique de surfacesminimales avec frontière libre et établissons de nouveaux résultats de comparaison avec sa varianteoù la fonctionnelle des surfaces minimales est remplacée par la variation totale. Nous généralisonsla notion de calibrabilité introduite par Caselles-Chambolle et Al. et construisons explicitementune solution duale pour le problème associé à la seconde fonctionnelle en utilisant un potentiellocalement Lipschitzien lié à la distance au cut-locus. La dernière partie de la thèse est consacrée auxalgorithmes d’optimisation de type primal-dual pour la recherche de points selle, en introduisant denouvelles variantes plus efficaces en précision et temps calcul. Nous avons en particulier introduit unevariante semi-implicite de la méthode d’Arrow-Hurwicz qui permet de réduire le nombre d’itérationsnécessaires pour obtenir une qualité satisfaisante des interfaces. Enfin nous avons traité la nondifférentiabilité structurelle des Lagrangiens utilisés à l’aide d’une méthode géométrique de projectionsur l’épigraphe offrant ainsi une alternative aux méthodes classiques de régularisation. / In this thesis, we study a general principle of convexification to treat certain non convex variationalproblems in Rd. Thanks to this principle we are able to enforce the powerful duality techniques andbring back such problems to primal-dual formulations in Rd+1, thus making efficient the numericalsearch of a global minimizer. A theory of duality and calibration fields is reformulated in the caseof linear-growth functionals. Under suitable assumptions, this allows us to revisit and extend anexclusion principle discovered by Visintin in the 1990s and to reduce the original problem to theminimization of a convex functional in Rd. This result is then applied successfully to a class offree boundary or multiphase problems that we treat numerically obtaining very accurate interfaces.On the other hand we apply the theory of calibrations to a classical problem of minimal surfaceswith free boundary and establish new results related to the comparison with its variant where theminimal surfaces functional is replaced by the total variation. We generalize the notion of calibrabilityintroduced by Caselles-Chambolle and Al. and construct explicitly a dual solution for the problemassociated with the second functional by using a locally Lipschitzian potential related to the distanceto the cut-locus. The last part of the thesis is devoted to primal-dual optimization algorithms forthe search of saddle points, introducing new more efficient variants in precision and computationtime. In particular, we experiment a semi-implicit variant of the Arrow-Hurwicz method whichallows to reduce drastically the number of iterations necessary to obtain a sharp accuracy of theinterfaces. Eventually we tackle the structural non-differentiability of the Lagrangian arising fromour method by means of a geometric projection method on the epigraph, thus offering an alternativeto all classical regularization methods.