Actuellement, les pratiques de Lean Management représentent un avantage compétitif pour la majorité des entreprises qui cherchent à améliorer leur performance dans un marché mondial très agressif. Le processus de mise en œuvre du Lean est très complexe ; il s'agit de se transformer en une nouvelle philosophie et de gérer l'entreprise - un changement de comportement. Ce document passe en revue la documentation relative aux pratiques de Lean Management et à l'incapacité de certaines entreprises à maintenir les résultats dans le temps, en particulier sur le décalage entre les objectifs du Lean et leurs efforts de mise en œuvre pour éliminer le gaspillage par l'amélioration des processus. La littérature a relevé plusieurs problèmes de gestion concernant ce problème, mais les principales raisons en sont, d'une part, une rupture des interfaces entre les aspects socio-techniques et, d'autre part, un besoin d'engagement réel de la part de la direction générale. Par conséquent, l'objectif de cet étude est de synthétiser et d'analyser ces difficultés Lean sur la base d'une pensée systémique dynamique et de proposer en plus, comme alternative à une proposition classique (linéarité) pour résoudre ces problèmes Leans, deux hypothèses : la contribution de la productivité qui réduit ces écarts, de manière plus globale ; en plus, sur l'approche de l'amélioration continue, qui permet de mesurer les"changements comportementaux" et encourage également la participation ; elle pose également le problème des performances chez des employés autonomes qui ont été documentés dans la documentation Lean / Currently, Lean Management Practices represents a competitive advantage for most companiestrying to raise their performance in a very aggressive global market. Lean’s implementationprocess is very complex; it means to transform into a new philosophy and managing the business- a behaviour change. This paper reviews the literature in relation to Lean managerial practicesand the incapacity for some companies to sustain the results over time; specifically about themisalignment among the Lean’s purposes with their implementation efforts to waste eliminationthrough the improvement of processes. The literature found several management issues regardingthis problem but the main reasons are firstly a breakdown interfaces between socio-technicalaspects and secondly, a need for real commitment from the top management. Consequently, thetarget of this paper is to synthesize and analyze those Lean difficulties based on dynamic systemthinking and, also, to propose two assumptions as an alternative to a conventional proposal(linearity) to solve this Leans’ problems: the contribution of productivity management whichnarrows these gaps, in a more holistic manner; in addition, based on the continuous improvementapproach as a metric to assess Lean’s "behaviour change" and also to encourage commitment; italso engages the performance dilemma throughout empowered workers that have beendocumented in the Lean literature

L'île de La Réunion, inscrite au patrimoine mondial de l'UNESCO pour la beauté de ses paysages et pour sa biodiversité exceptionnelle, va accueillir dans les prochaines années une population de plus en plus importante. Les documents de planification réglementant l'évolution de l'urbanisation dans l'île doivent donc en tenir compte afin de gérer au mieux l'évolution et la structuration du foncier local. Dans un contexte de recherche pluridisciplinaire, l'enjeu de nos travaux est donc de proposer des méthodes et outils facilitant les interactions entre les informaticiens et les thématiciens qui apportent leur expertise à la construction de modèles de simulation dédiés à la prospective territoriale et de faire en sorte que ces modèles puissent fournir des pistes de réflexions aux décideurs confrontés aux choix de demain afin d'aider à mettre en place une organisation des territoires qui soit la plus cohérente possible. Après avoir participé à la réalisation de DS, un modèle qui permet de simuler conjointement l'évolution de la population et celle du Mode d'Occupation des Sols à La Réunion, nous nous sommes focalisés sur l'étude de l'émergence, une notion qui apparaît dans nos simulations en environnement spatialisé, mais pour laquelle il n'existe pas d'outils génériques permettant de la manipuler. Nous l'avons définie comme étant une méta-connaissance, ce qui nous a permis de proposer une méthodologie de conception de simulations et un cadre formel qui ont abouti à la mise en place de structures émergentes dans la plateforme de simulation GEAMAS-NG. Celle-ci, ou les entités de la SOA elles-mêmes, peuvent ainsi prendre conscience de l'émergence de phénomènes et les matérialiser en ayant recours à la connaissance des thématiciens. Nous avons alors montré l'intérêt de ces propositions en expérimentant la réification de phénomènes émergents observés dans DS. / Reunion Island will be hosting in the coming years an increasingly important population. Documents dedicated to land-use planning should take this into account to better manage the evolutions of the island. In the context of multidisciplinary research, the challenge of our work is to propose methods and tools to facilitate interactions between computer scientists and thematicians who bring their expertise to the construction of simulation models dedicated to territorial prospective. We co-constructed DS, an agent-based model that simulates the evolutions of the population and of the land-use on the island, and then we focused on the study of emergence, a concept that appears in our simulations in spatial environment. We defined it as a meta-knowledge; and that enabled us to propose a modelling methodology and a formal framework that lead to the introduction of emerging structures in the Agent-Based Simulation platform GEAMAS-NG. The systems, or the entities of the models themselves, can now be aware of the emergence of phenomena and can materialize these phenomena by using the knowledge of thematicians. We finally demonstrated the usefulness of these proposals by experimenting the reification of an emergent phenomenon observed in DS.

Les systèmes de prévision des crues décrivent les transformations pluie-débit en se basant sur des représentations simplifiées. Ces représentations modélisent les processus physiques impliqués avec des descriptions empiriques, ou basées sur des équations de la mécanique classique. Les performances des modèles actuels de prévision des crues sont affectées par différentes incertitudes liées aux approximations et aux paramètres du modèle, aux données d’entrée et aux conditions initiales du bassin versant. La connaissance de ces incertitudes permet aux décideurs de mieux interpréter les prévisions et constitue une aide à la décision lors de la gestion de crue. L’analyse d’incertitudes dans les modèles hydrologiques existants repose le plus souvent sur des simulations de Monte-Carlo (MC). La mise en œuvre de ce type de techniques requiert un grand nombre de simulations et donc un temps de calcul potentiellement important. L'estimation des incertitudes liées à la modélisation hydrologique en temps réel reste donc une gageure. Dans ce projet de thèse, nous développons une méthodologie de prévision des crues basée sur les réseaux Bayésiens (RB). Les RBs sont des graphes acycliques dans lesquels les nœuds correspondent aux variables caractéristiques du système modélisé et les arcs représentent les dépendances probabilistes entre ces variables. La méthodologie présentée propose de construire les RBs à partir des principaux facteurs hydrologiques contrôlant la génération des crues, en utilisant à la fois les observations disponibles de la réponse du système et les équations déterministes décrivant les processus concernés. Elle est conçue pour prendre en compte la variabilité temporelle des différentes variables impliquées. Les dépendances probabilistes entre les variables (paramètres) peuvent être spécifiées en utilisant des données observées, des modèles déterministes existants ou des avis d’experts. Grâce à leurs algorithmes d’inférence, les RBs sont capables de propager rapidement, à travers le graphe, différentes sources d'incertitudes pour estimer leurs effets sur la sortie du modèle (ex. débit d'une rivière). Plusieurs cas d’études sont testés. Le premier cas d’étude concerne le bassin versant du Salat au sud-ouest de la France : un RB est utilisé pour simuler le débit de la rivière à une station donnée à partir des observations de 3 stations hydrométriques localisées en amont. Le modèle présente de bonnes performances pour l'estimation du débit à l’exutoire. Utilisé comme méthode inverse, le modèle affiche également de bons résultats quant à la caractérisation de débits d’une station en amont par propagation d’observations de débit sur des stations en aval. Le deuxième cas d’étude concerne le bassin versant de la Sagelva situé en Norvège, pour lequel un RB est utilisé afin de modéliser l'évolution du contenu en eau de la neige en fonction des données météorologiques disponibles. Les performances du modèle sont conditionnées par les données d’apprentissage utilisées pour spécifier les paramètres du modèle. En l'absence de données d'observation pertinentes pour l’apprentissage, une méthodologie est proposée et testée pour estimer les paramètres du RB à partir d’un modèle déterministe. Le RB résultant peut être utilisé pour effectuer des analyses d’incertitudes sans recours aux simulations de Monte-Carlo. Au regard des résultats enregistrés sur les différents cas d’études, les RBs se révèlent utiles et performants pour une utilisation en support d’un processus d'aide à la décision dans le cadre de la gestion du risque de crue. / Flood forecasting describes the rainfall-runoff transformation using simplified representations. These representations are based on either empirical descriptions, or on equations of classical mechanics of the involved physical processes. The performances of the existing flood predictions are affected by several sources of uncertainties coming not only from the approximations involved but also from imperfect knowledge of input data, initial conditions of the river basin, and model parameters. Quantifying these uncertainties enables the decision maker to better interpret the predictions and constitute a valuable decision-making tool for flood risk management. Uncertainty analysis on existing rainfall-runoff models are often performed using Monte Carlo (MC)- simulations. The implementation of this type of techniques requires a large number of simulations and consequently a potentially important calculation time. Therefore, quantifying uncertainties of real-time hydrological models is challenging. In this project, we develop a methodology for flood prediction based on Bayesian networks (BNs). BNs are directed acyclic graphs where the nodes correspond to the variables characterizing the modelled system and the arcs represent the probabilistic dependencies between these variables. The presented methodology suggests to build the RBs from the main hydrological factors controlling the flood generation, using both the available observations of the system response and the deterministic equations describing the processes involved. It is, thus, designed to take into account the time variability of different involved variables. The conditional probability tables (parameters), can be specified using observed data, existing hydrological models or expert opinion. Thanks to their inference algorithms, BN are able to rapidly propagate, through the graph, different sources of uncertainty in order to estimate their effect on the model output (e.g. riverflow). Several case studies are tested. The first case study is the Salat river basin, located in the south-west of France, where a BN is used to simulate the discharge at a given station from the streamflow observations at 3 hydrometric stations located upstream. The model showed good performances estimating the discharge at the outlet. Used in a reverse way, the model showed also satisfactory results when characterising the discharges at an upstream station by propagating back discharge observations of some downstream stations. The second case study is the Sagelva basin, located in Norway, where a BN is used to simulate the accumulation of snow water equivalent (SWE) given available weather data observations. The performances of the model are affected by the learning dataset used to train the BN parameters. In the absence of relevant observation data for learning, a methodology for learning the BN-parameters from deterministic models is proposed and tested. The resulted BN can be used to perform uncertainty analysis without any MC-simulations to be performed in real-time. From these case studies, it appears that BNs are a relevant decisionsupport tool for flood risk management.

Pour mettre en relation les propriétés rhéologiques et la structure méso/microscopique d'un système modèle ou complexe, l'utilisation de la rhéo-optique est indispensable. Nous avons donc développé une cellule d'observation sous cisaillement adaptée à la microscopie confocale. Ce dispositif, breveté et nommé RheOptiCAD®, permet le cisaillement contrôlé d'un échantillon quelconque placé entre 2 plans parallèles en translation. Grâce à un système à dépression, la mise en place de l'échantillon est simple et rapide tout en assurant des propriétés optiques répétables et reproductibles (planéité, parallélisme). Par ailleurs, la température au sein de l'échantillon peut être régulée de façon à imposer une contrainte thermique, cause de nombreuses modifications de la mésostructure d'un système alimentaire. La cellule d'observation sous cisaillement permet donc de suivre l'évolution et la dynamique des changements de structures conséquences d'un traitement thermo-mécanique imposé. Un logiciel de pilotage et d'acquisition des données a été développé pour rendre son utilisation plus conviviale. La validation du fonctionnement de l'outil et de ses fonctionnalités a tout d'abord été réalisée sans échantillon puis à l'aide d'un système modèle contenant des particules fluorescentes dont le mouvement était suivi. Par la suite, dans le but de tester les potentialités de ce nouvel outil tout en développant la méthodologie de son utilisation, et en particulier l'équilibre entre propriétés optiques et mécaniques des échantillons, nous avons travaillé avec de la pâte de farine. Ce système alimentaire bien connu et maîtrisé d'un point de vue rhéologique au laboratoire présente des caractéristiques intéressantes dans ce cadre. L'évolution du réseau de gluten au cours d'un cisaillement oscillatoire en fonction de la formulation de la pâte a été étudiée. Grâce à une analyse d'image basée sur la morphologie mathématique, nous avons pu mettre en évidence des changements de structures au cours du temps. De même, à l'aide des capacités thermiques de la cellule de cisaillement, nous avons étudié le positionnement et le mouvement des lipides endogènes à l'interface air-protéine lors de la fermentation. Notre cellule d'observation sous cisaillement constitue donc un nouvel outil de caractérisation dynamique de systèmes complexes couplant rhéologie et microscopie. Son optimisation principale réside dans la mise en place d'un capteur de force, mesurant les contraintes mises en jeu lors des déformations imposées. / Rheo-optic is a recent technique which can be used to create links between rheological properties and meso/microsctructures of model or complex (food) systems. A novel rheo-optical shearing device was designed for studying this relationships within complex food systems. The device has been build to be adapted on an inverted confocal microscope. Specifications of the shear cell are: a) a controlled translational shear between 2 parallel plates with three different motion modes (continuous, oscillatory, strain jump); b) a thermal control; and c) an observation on an inverted confocal microscope. Due to a vacuum system, the set up of an experiment is easy and fast ensuring reproducible optical properties (planarity, parallelism). Temperature, responsible of numerous modifications of structures in a food matrix, is also controlled. A piloting software allows an easy use of the shear cell. Validation of the motion modes has been carried out using a microgel, containing fluorescent probes (spheres) and tracking some of the particles. Next, in order to test and develop methods of observation under shear, taking into account the optical-mechanical balance, bread dough observation has been performed. Well known and described in the lab, bread dough is a dispersion of air bubbles and starch granules in a gluten network. Evolution on this gluten network depending on the formulation of the bread dough has been studied under oscillatory shearing. The composition effect on the microstructure and its evolution were observed and will be commented. Image analysis based on grayscale mathematical morphology has been carried out in order to try to quantify the rheological properties and microstructures. Finally, by a controlled increase of temperature, the growth of an air bubble in bread dough containing yeast was followed during proofing. The influence and the disposition of fat globules at the bubble air-protein interface along this growing process were followed. Thanks to the rheo-optical device, images of microstructures obtained under controlled shear are compared to their rheological behaviour.

Les enjeux planétaires du développement durable et du réchauffement climatique remettent en cause le fonctionnement de nos sociétés. La transition énergétique se traduit notamment par l’augmentation de l’utilisation des énergies renouvelables et constitue ainsi le contexte de nos travaux. Dans le cadre d’un contrat CIFRE avec l’entreprise VALOREM, nous proposons des méthodologies pour le développement et la conception d’installations renouvelables soutenables. L’approche d’Ingénierie Des Modèles mise en place nous permet de garder une démarche scientifique cohérente entre une vue très générale du contexte d’insertion des renouvelables et une vision plus technique et opérationnelle pour le développement d’installations renouvelables. Dans un premier temps, nous avons décrit de manière macroscopique le contexte d’insertion des énergies renouvelables en France, en proposant un cadre conceptuel pour le développement de projet d’aménagement du territoire. Il permet l’analyse des systèmes sociotechniques en transition grâce à une formalisation UML2 d’une part, et une méthode d’évaluation de la durabilité d’un processus d’autre part. Nous avons illustré ce cadre sur deux cas d’étude : le succès du développement de la centrale hydro-éolienne de l’île d’El Hierro dans les Canaries et l’échec du barrage de Sivens dans le Tarn en France. Par comparaison, nous montrons que l’implication des parties prenantes sur l’ensemble du processus de développement est cruciale pour le succès d’un projet. Nous avons ensuite proposé une implémentation logicielle permettant le suivi et le pilotage de projet d’aménagement du territoire en temps réel à l’aide d’outils d’intelligence artificielle. Conscient des limites opérationnelles de notre proposition, nous avons décidé de changer de point de vue pour adopter une vision plus technique sur les énergies renouvelables. La transition énergétique fait évoluer le système électrique français d’un modèle centralisé vers un modèle décentralisé, ce qui amène VALOREM à reconsidérer ses propositions techniques de production d’énergies renouvelables. Entre autre, l’entreprise tend vers l’hybridation de ses centrales et constate le besoin d’un outil de dimensionnement en phase d’avant-projet, afin d’identifier les opportunités d’activités. Nous faisons donc la proposition d’un prototype logiciel qui se veut flexible en termes de modélisation des centrales de production hybrides renouvelables et en termes de définition des objectifs et contraintes à satisfaire. Il est codé en C++ et basé sur des modèles d’optimisation linéaires. Nous illustrons son opérationnalité sur quatre cas d’étude : deux orientés marché électrique et deux en autoconsommation (l’un isolé, l’autre connecté au réseau). Les résultats des études montrent que les coûts de stockage batterie sont trop élevés pour pouvoir envisager l’installation de batteries dans les centrales hybrides renouvelables. Les systèmes en autoconsommation connectés au réseau semblent être plus prometteurs. Enfin, nous terminons par une analyse environnementale de différentes sources de production électrique afin de déterminer celles vers lesquelles la transition énergétique doit être axée. Les résultats montrent que la biomasse, l’éolien et le PV au sol sont les moyens de production les plus propres. La conclusion de la thèse propose une réflexion ascendante en complexité vers l’implication des acteurs dans les modèles de processus de développement territoriaux. / Sustainable growth issues and climate change imply to shift our society organisation towards new paradigms. In the energy sector, it is illustrated by an energy transition via renewables. Within this context and the support of the company VALOREM, we propose a methodology for the sustainable development and design of renewable infrastructures. By using a Model Driven Engineering approach, we were able to adopt several viewpoints on renewable energy issues, while keeping a consistent scientific approach. First of all, from a holistic stance, we developed a conceptual framework for land settlement development process. Thanks to its modelling with UML2 and a methodology to assess the sustainability of processes, it can be used to analyse the trajectory of sociotechnical systems. The framework was tested on two case studies: the success of El Hierro hydrowind power plant implementation in Canary Islands and the failure of the Sivens dam project in the south of France. Their comparison shows that implication of stakeholders at all stages of the development process is crucial for the success of the project. We further propose to implement the conceptual framework as a software tool for the monitoring of land settlement project in real time, with the help of artificial intelligence concepts. However, this proposition was deemed not in line with Valorem’s priorities and we decided to focus on a technical instantiation of the framework for Hybride Renewable Energy Systems (HRES), which design and operation are part of VALOREM’s activity. As, the energy transition implies to change the electrical system model from a centralized to a decentralized one, there exists a need for a grassroots sizing tool for identifying new HRES opportunities. A software prototype for the grassroot optimal design of HRES was developed, offering flexibility in terms of modelling, setting constraints and defining objective functions. It is coded in C++ and based on linear programming model. We showed its relevance on four HRES case studies: two related to electricity market and two for self-supply of energy (one isolated and the other grid-connected). The results show that battery costs are too high to be profitable in HRES and that grid-connected systems seems to be a good alternative. Finally, we carried out an environmental impact assessment of several sources of electricity production to find out which are the “cleaner” ones. The results show that biomass, wind turbines and ground PV are those that should be considered for a sustainable energy transition. In the conclusion, we step back to a more complex viewpoint and develop perspectives to model human interactions within territory development process.

Complex Systems-of-Systems, which form the subject matter of the research presented here, may be considered to be systems that are constituted of several components, each of which could themself be regarded as being a complete system in its own right. These component systems are typically largely managed independently of one another. They each have their own intrinsic goals which they are capable of fulfilling autonomously. These goals are additional to any to which they might contribute as part of the overall System-of-Systems that is being investigated. Systems-of-Systems are epitomised by features such as the prevalence of complex webs of mutual interdependency amongst their component systems, the involvement of a broad diversity of stakeholders and the participation of a number of independent decision-makers, each of whom competes with the others for strictly limited resources. Furthermore, the components of Systems-of-Systems may often be found to be widely scattered across an extensive geographical region. Sometimes such systems may even span global distances. Complex Systems-of-Systems of such a nature would appear to be becoming ever more prevalent in the increasingly interconnected and ever shrinking world which is engendered, in particular, by the proliferation of modern information and communication technology. Systems-of-Systems may arise in both commercial and public sector contexts, hence commercial organisations and governments, alike, confront the daunting challenge of planning, establishing and maintaining novel systems of this type. However, it is to be anticipated that significant difficulties might be encountered in applying existing systems engineering methods, which were designed for dealing with far more monolithic types of system, to the treatment of Systems-of- Systems. Instead, the successful achievement of acceptable resolutions to the various Systems-of-Systems issues and conundrums with which the responsible planners can be expected to be confronted would appear to demand the application of sophisticated distributed decision aids, in order to alleviate the the intolerable burden which the decision-makers would otherwise be forced to endure. Such aids need to be able to take a genuinely fair and unbiased account of the interests of a multitude of stakeholders. At the same time, they must be able to satisfactorily accommodate the enforcement of a range of highly involved and intricate constraints upon the various alternative interventions which might potentially be contemplated. In order to underpin the development of decision aids of the kind demanded, it has been sought to establish some basic theoretical foundations for Systems-of-Systems. This body of theory has then been applied to the identification of the methods which might be most suitable for the treatment of Systems-of-Systems questions. Finally, some decision-support tools which are intended to facilitate the implementation of the methods that have been recommended have been described. Any attempt to dictate a single â??bestâ?? intervention to the decision-makers would generally be totally unacceptable. As a result, the objective which has been pursued has been to seek to filter out and identify a relatively sparing number of what would appear to be the most reasonable alternatives from some much more profuse range of possibilities. These may then be afforded more intensive scrutiny by the pertinent decision-makers.

Complex systems are those which exhibit one or more of the following inter-related behaviours: 1. Nonlinear behaviour: the component parts do not act in linear ways, that is the superposition of the actions of the parts is not the output of the system. 2. Emergent behaviour: the output of the system may be inexpressible in terms of the rules or equations of the component parts. 3. Self-organisation: order appears from the chaotic interactions of individuals and the rules they obey. 4. Layers of description: in which a rule may apply at some higher levels of description but not at lower layers. 5. Adaptation: in which the environment becomes encoded in the rules governing the structure and/or behaviour of the parts (in this case strictly agents) that undergo selection in which those that are by some measure better become more numerous than those that are not as “fit”. A single cell is a complex system: we cannot explain all of its behaviour as simply the sum of its parts. Similarly, DNA structures, social networks, cancers, the brain, and living beings are intricate complex systems. This thesis tackles all of these topics from a complex systems approach. I have skirted some of the philosophical issues of complex systems and mainly focussed on appropriate tools to analyse these systems, addressing important questions such as: • What is the best way to extract information from DNA? • How can we model and analyse mutations in DNA? • Can we determine the likely spread of both viruses and ideas in social networks? • How can we model the growth of cancer? • How can we model and analyse interactions between genes in such living systems as the fruit fly, cancers, and humans? • Can complex systems techniques give us some insight into the human brain? / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1290759 / Thesis (Ph.D.)-- School of Electrical and Electronic Engineering, 2007

<p>In every living cell, millions of different types of molecules constantly interact and react chemically in a complex system that can adapt to fluctuating environments and extreme conditions, living to survive and reproduce itself. The information required to produce these components is stored in the genome, which is copied in each cell division and transferred and mixed with another genome from parent to child. The regulatory mechanisms that control biological systems, for instance the regulation of expression levels for each gene, has evolved so that global robustness and ability to survive under harsh conditions is a strength, at the same time as biological tasks on a detailed molecular level must be carried out with good precision and without failures. This has resulted in systems that can be described as a hierarchy of levels of complexity: from the lowest level, where molecular mechanisms control other components at the same level, to pathways of coordinated interactions between components, formed to carry out particular biological tasks, and up to large-scale systems consisting of all components, connected in a network with a topology that makes the system robust and flexible. This thesis reports on work that model and analyze complex biological systems, and the signals and noise that regulate them, at all different levels of complexity. Also, it shows how signals are transduced vertically from one level to another, as when a single mutation can cause errors in low level mechanisms, disrupting pathways and create systemwide imbalances, such as in type 2 diabetes. The advancement of our knowledge of biological systems requires both that we go deeper and towards more detail, of single molecules in single cells, as well as taking a step back to understand the organisation and dynamics in the large networks of all components, and unite the different levels of complexity.</p>

In this dissertation, I propose that network theory offers a useful frame for informing mathematics education. Mathematical understanding, like the discipline of formal mathematics within which it is subsumed, possesses attributes characteristic of complex systems. As the techniques of network theorists are often used to explore such forms, a network model provides a novel and productive way to interpret individual comprehension of mathematics. A network structure for mathematical understanding can be found in cognitive mechanisms presented in the theory of embodied mathematics described by Lakoff and Nez. Specifically, conceptual domains are taken as the nodes of a network and conceptual metaphors as the connections among them. Examination of this metaphoric network of mathematics reveals the scale-free topology common to complex systems. Patterns of connectivity in a network determine its dynamic behavior. Scale-free systems like mathematical understanding are inherently vulnerable, for cascading failures, where misunderstanding one concept can lead to the failure of many other ideas, may occur. Adding more connections to the metaphoric network decreases the likelihood of such a collapse in comprehension. I suggest that an individuals mathematical understanding may be made more robust by ensuring each concept is developed using metaphoric links that supply patterns of thought from a variety of domains. Ways of making this a focus of classroom instruction are put forth, as are implications for curriculum and professional development. A need for more knowledge of metaphoric connections in mathematics is highlighted. To exemplify how such research might be carried out, and with the intent of substantiating ideas presented in this dissertation, I explore a small part of the proposed metaphoric network around the concept of EXPONENTIATION. Using collaborative discussion, individual interviews and literature, a search for representations that provide varied ways of making sense of EXPONENTIATION is carried out. Examination of the physical and mathematical roots of these conceptualizations leads to the identification of domains that can be linked to EXPONENTIATION.

In every living cell, millions of different types of molecules constantly interact and react chemically in a complex system that can adapt to fluctuating environments and extreme conditions, living to survive and reproduce itself. The information required to produce these components is stored in the genome, which is copied in each cell division and transferred and mixed with another genome from parent to child. The regulatory mechanisms that control biological systems, for instance the regulation of expression levels for each gene, has evolved so that global robustness and ability to survive under harsh conditions is a strength, at the same time as biological tasks on a detailed molecular level must be carried out with good precision and without failures. This has resulted in systems that can be described as a hierarchy of levels of complexity: from the lowest level, where molecular mechanisms control other components at the same level, to pathways of coordinated interactions between components, formed to carry out particular biological tasks, and up to large-scale systems consisting of all components, connected in a network with a topology that makes the system robust and flexible. This thesis reports on work that model and analyze complex biological systems, and the signals and noise that regulate them, at all different levels of complexity. Also, it shows how signals are transduced vertically from one level to another, as when a single mutation can cause errors in low level mechanisms, disrupting pathways and create systemwide imbalances, such as in type 2 diabetes. The advancement of our knowledge of biological systems requires both that we go deeper and towards more detail, of single molecules in single cells, as well as taking a step back to understand the organisation and dynamics in the large networks of all components, and unite the different levels of complexity.