Cette thèse se penche sur les problèmes de décisions devant être prises de manière séquentielle au sein d'un environnement aléatoire. Lors de chaque étape d'un tel problème décisionnel, une alternative doit être sélectionnée parmi un ensemble d'alternatives. Chaque alternative possède un gain moyen qui lui est propre et lorsque l'une d'elles est sélectionnée, celle-ci engendre un gain aléatoire. La sélection opérée peut suivre deux types d'objectifs.<p>Dans un premier cas, les tests viseront à maximiser la somme des gains collectés. Un juste compromis doit alors être trouvé entre l'exploitation et l'exploration. Ce problème est couramment dénommé dans la littérature scientifique "multi-armed bandit problem".<p>Dans un second cas, un nombre de sélections maximal est imposé et l'objectif consistera à répartir ces sélections de façon à augmenter les chances de trouver l'alternative présentant le gain moyen le plus élevé. Ce deuxième problème est couramment repris dans la littérature scientifique sous l'appellation "selecting the best".<p>La sélection de type gloutonne joue un rôle important dans la résolution de ces problèmes de décision et opère en choisissant l'alternative qui s'est jusqu'ici montrée optimale. Or, la nature généralement aléatoire de l'environnement rend incertains les résultats d'une telle sélection. <p>Dans cette thèse, nous introduisons une nouvelle quantité, appelée le "gain espéré d'une action gloutonne". Sur base de quelques propriétés de cette quantité, de nouveaux algorithmes permettant de résoudre les deux problèmes décisionnels précités seront proposés.<p>Une attention particulière sera ici prêtée à l'application des techniques présentées au domaine de la sélection de modèles en l'apprentissage artificiel supervisé. <p>La collaboration avec le service d'anesthésie de l'Hôpital Erasme nous a permis d'appliquer les algorithmes proposés à des données réelles, provenant du milieu médical. Nous avons également développé un système d'aide à la décision dont un prototype a déjà été testé en conditions réelles sur un échantillon restreint de patients. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

A swarm intelligence system is a type of multiagent system with the following distinctive characteristics: (i) it is composed of a large number of agents, (ii) the agents that comprise the system are simple with respect to the complexity of the task the system is required to perform, (iii) its control relies on principles of decentralization and self-organization, and (iv) its constituent agents interact locally with one another and with their environment. <p><p>Interactions among agents, either direct or indirect through the environment in which they act, are fundamental for swarm intelligence to exist; however, there is a class of interactions, referred to as "interference", that actually blocks or hinders the agents' goal-seeking behavior. For example, competition for space may reduce the mobility of robots in a swarm robotics system, or misleading information may spread through the system in a particle swarm optimization algorithm. One of the most visible effects of interference in a swarm intelligence system is the reduction of its efficiency. In other words, interference increases the time required by the system to reach a desired state. Thus, interference is a fundamental problem which negatively affects the viability of the swarm intelligence approach for solving important, practical problems.<p><p>We propose a framework called "incremental social learning" (ISL) as a solution to the aforementioned problem. It consists of two elements: (i) a growing population of agents, and (ii) a social learning mechanism. Initially, a system under the control of ISL consists of a small population of agents. These agents interact with one another and with their environment for some time before new agents are added to the system according to a predefined schedule. When a new agent is about to be added, it learns socially from a subset of the agents that have been part of the system for some time, and that, as a consequence, may have gathered useful information. The implementation of the social learning mechanism is application-dependent, but the goal is to transfer knowledge from a set of experienced agents that are already in the environment to the newly added agent. The process continues until one of the following criteria is met: (i) the maximum number of agents is reached, (ii) the assigned task is finished, or (iii) the system performs as desired. Starting with a small number of agents reduces interference because it reduces the number of interactions within the system, and thus, fast progress toward the desired state may be achieved. By learning socially, newly added agents acquire knowledge about their environment without incurring the costs of acquiring that knowledge individually. As a result, ISL can make a swarm intelligence system reach a desired state more rapidly. <p><p>We have successfully applied ISL to two very different swarm intelligence systems. We applied ISL to particle swarm optimization algorithms. The results of this study demonstrate that ISL substantially improves the performance of these kinds of algorithms. In fact, two of the resulting algorithms are competitive with state-of-the-art algorithms in the field. The second system to which we applied ISL exploits a collective decision-making mechanism based on an opinion formation model. This mechanism is also one of the original contributions presented in this dissertation. A swarm robotics system under the control of the proposed mechanism allows robots to choose from a set of two actions the action that is fastest to execute. In this case, when only a small proportion of the swarm is able to concurrently execute the alternative actions, ISL substantially improves the system's performance. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

A concept for the mathematical modeling of multiple repairable equipment and logistic systems (MREAL systems) is developed; These systems consist of multiple populations of repairable equipment, and their associated design, procurement, maintenance, and supply support. MREAL systems present management and design problems which parallel the·management and design of multiple, consumable item inventory systems. However, the MREAL system is more complex since it has a repair component. The MREAL system concept is described in a classification hierarchy which attempts to categorize the components of such systems. A specific mathematical model (MREAL1) is developed for a subset of these components. Included in MREAL1 are representations of the equipment reliability and maintainability design problem, the maintenance capacity problem, the retirement age problem, and the population size problem, for each of the multiple populations. MREAL1 models the steady state stochastic behavior of the equipment repair facilities using an approximation which is based upon the finite source, multiple server queuing system. System performance measures included in MREAL1 are: the expected MREAL total system life cycle cost (including a shortage cost penalty); the steady state expected number of shortages; the probability of catastrophic failure in each equipment population; and two budget based measures of effectiveness. Two optimization methods are described for a test problem developed for MREAL1. The first method computes values of the objective function and the constraints for a specified subset of the solution space. The best feasible solution found is recorded. This method can also examine all possible solutions, or can be used in a manual search. The second optimization method performs an exhaustive enumeration. of the combinatorial programming portion of MREAL1, which represents equipment design. For each enumerated design combination, an attempt is made to find the optimal solution to the remaining nonlinear discrete programming problem. A sequential unconstrained minimization technique is used which is based on an augmented Lagrangian penalty function adapted to the integer nature of MREAL1. The unconstrained minimization is performed by a combination of Rosenbrock's search technique, the steepest descent method, and Fibonacci line searches, adapted to the integer nature of the search. Since the model contains many discrete local minima, the sequential unconstrained minimization is repeated from different starting solutions, based upon a heuristic selection procedure. A gradient projection method provides the termination criteria for each unconstrained minimization. / Ph. D.

The recent global financial crisis, regulatory and compliance requirements placed on organisations, and the need for scientific research in the project portfolio management discipline were factors that motivated this research. The interest and contribution to the body of knowledge in project portfolio management has been growing significantly in recent years, however, there still appears to be a misalignment between literature and practice. A particular area of concern is the decision-making, during the management of the portfolio, regarding which projects to accelerate, suspend, or terminate. A lack of determining the individual and cumulative contribution of projects to strategic objectives leads to poorly informed decisions that negate the positive effect that project portfolio management could have in an organisation. The focus of this research is, therefore, aimed at providing a mechanism to determine the individual and cumulative contribution of projects to strategic objectives so that the right decisions can be made regarding those projects. This thesis begins with providing a context for project portfolio management by confirming a definition and providing a theoretical background through related theories. An investigation into the practice of project portfolio management then provides insight into the alignment between literature and practice and confirms the problem that needed to be addressed. A conceptual model provides a solution to the problem of determining the individual and cumulative contribution of projects to strategic objectives. The researcher illustrates how the model can be extended before verifying and validating the conceptual model. Having the ability to determine the contributions of projects to strategic objectives affords decision makers the opportunity to conduct what-if scenarios, enabled through the use of dashboards as a visualization technique, in order to test the impact of their decisions before committing them. This ensures that the right decisions regarding the project portfolio are made and that the maximum benefit regarding the strategic objectives is achieved. This research provides the mechanism to enable better-informed decision- making regarding the project portfolio. / Computing / D. Phil. (Computer science)

This project extended COLLDESIGN, an interactive collaborative modeling tool that was developed by Mr. Hara Totapally. The initial version included a collaborative framework comprised of configurable client and server components. This project accomplished a complete implementation of the Class Diagram view. In addition, extending the framework, text messaging and audio conferencing features have been implemented to allow for real-time textual and audio communication between team members working on a particular project. VideoClient is the GUI of the application.

The recent global financial crisis, regulatory and compliance requirements placed on organisations, and the need for scientific research in the project portfolio management discipline were factors that motivated this research. The interest and contribution to the body of knowledge in project portfolio management has been growing significantly in recent years, however, there still appears to be a misalignment between literature and practice. A particular area of concern is the decision-making, during the management of the portfolio, regarding which projects to accelerate, suspend, or terminate. A lack of determining the individual and cumulative contribution of projects to strategic objectives leads to poorly informed decisions that negate the positive effect that project portfolio management could have in an organisation. The focus of this research is, therefore, aimed at providing a mechanism to determine the individual and cumulative contribution of projects to strategic objectives so that the right decisions can be made regarding those projects. This thesis begins with providing a context for project portfolio management by confirming a definition and providing a theoretical background through related theories. An investigation into the practice of project portfolio management then provides insight into the alignment between literature and practice and confirms the problem that needed to be addressed. A conceptual model provides a solution to the problem of determining the individual and cumulative contribution of projects to strategic objectives. The researcher illustrates how the model can be extended before verifying and validating the conceptual model. Having the ability to determine the contributions of projects to strategic objectives affords decision makers the opportunity to conduct what-if scenarios, enabled through the use of dashboards as a visualization technique, in order to test the impact of their decisions before committing them. This ensures that the right decisions regarding the project portfolio are made and that the maximum benefit regarding the strategic objectives is achieved. This research provides the mechanism to enable better-informed decision- making regarding the project portfolio. / Computing / D. Phil. (Computer science)

In this thesis, we use evolutionary robotics techniques to automatically design and synthesise<p>behaviour for groups of simulated and real robots. Our contribution will be on<p>the design of non-trivial individual and collective behaviour; decisions about solitary or<p>social behaviour will be temporal and they will be interdependent with communicative<p>acts. In particular, we study time-based decision-making in a social context: how the<p>experiences of robots unfold in time and how these experiences influence their interaction<p>with the rest of the group. We propose three experiments based on non-trivial real-world<p>cooperative scenarios. First, we study social cooperative categorisation; signalling and<p>communication evolve in a task where the cooperation among robots is not a priori required.<p>The communication and categorisation skills of the robots are co-evolved from<p>scratch, and the emerging time-dependent individual and social behaviour are successfully<p>tested on real robots. Second, we show on real hardware evidence of the success of evolved<p>neuro-controllers when controlling two autonomous robots that have to grip each other<p>(autonomously self-assemble). Our experiment constitutes the first fully evolved approach<p>on such a task that requires sophisticated and fine sensory-motor coordination, and it<p>highlights the minimal conditions to achieve assembly in autonomous robots by reducing<p>the assumptions a priori made by the experimenter to a functional minimum. Third, we<p>present the first work in the literature to deal with the design of homogeneous control<p>mechanisms for morphologically heterogeneous robots, that is, robots that do not share<p>the same hardware characteristics. We show how artificial evolution designs individual<p>behaviours and communication protocols that allow the cooperation between robots of<p>different types, by using dynamical neural networks that specialise on-line, depending on<p>the nature of the morphology of each robot. The experiments briefly described above<p>contribute to the advancement of the state of the art in evolving neuro-controllers for<p>collective robotics both from an application-oriented, engineering point of view, as well as<p>from a more theoretical point of view. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

Thesis (MSc)--University of Stellenbosch, 2006. / ENGLISH ABSTRACT: Rural development problems faced by the impoverished communities in the Transkei, South Africa, are numerous, and environmental degradation has already taken much of its toll. By working at a micro-catchment-level both the socio-economic and biophysical appreciation of the land resources were captured as encapsulated in the concept of resource management domains. Participatory decision-making allowed functional land use goals and evaluation criteria to be incorporated into computerised multi-criteria evaluation and multi-objective land use allocation models in order to reach an idealised or more sustainable land use situation. In the execution of the decision-making process seven procedural steps were followed, which are discussed in detail and applied in the case study. Synthesis of the results emphasised the envisaged rural planning potential of the methods used. / AFRIKAANSE OPSOMMING: In terme van plattelandse ontwikkeling staar talle probleme die behoeftige gemeenskappe van Transkei, Suid-Afrika, in die gesig en omgewingsdegradering neem ongehinderd sy tol. Deur op ‘n mikro-opvangsgebied vlak te werk kon beide die sosio-ekonomiese en biofisiese waarde van die gebied se hulpbronne bepaal word en uitgebeeld word in hulpbron bestuursdomeine. Deur deelnemende besluitneming is funksionele grondgebruiksdoelwitte en evaluasie kriteria gebruik in gerekenariseerde meervoudige kriteria evaluering en veeldoelige grondgebruiksaanwysingsmodelle ten einde die ideale of ‘n meer volhoubare grondgebruik situasie te verkry. Vir die uitvoering van die besluitnemingsproses is van sewe opeenvolgende stappe gebruik gemaak en die uitvoering daarvan word in diepte bespreek in hierdie gevallestudie. Sintese van die resultate het die potensiaal van hierdie beoogde landelike beplanningsmetodes beklemtoon.