L'électricité joue un rôle de plus en plus important dans notre société. En effet, nous nous dirigeons vers l'ère du "tout électrique". Les besoins évoluant, il est indispensable de repenser la manière dont l'électricité est produite et distribuée. Cela introduit le concept de Smart Grid. Le Smart Grid est un concept de réseau électrique capable de supporter de manière autonome et intelligente les changements et pannes qui pourraient survenir dans un réseau. Cela répond directement au fait que de part la nature fortement distribuée et l'imprédictibilité de l'environnement (météo, ...), ces événements sont imprévisibles. Pour cela, cette thèse propose un cadre applicatif (framework) innovant basé sur les multi-agents ainsi que la conception et l'implémentation de comportements coopératifs pour résoudre deux problémes courants dans les réseaux électriques: l'analyse des flux de puissance et l'estimation d'état. Ces problèmes ont été abordés avec l'approche des Systèmes Multi-Agent Adaptatifs. Ces systèmes sont efficaces pour résoudre des problèmes complexes et ont la capacité d'adapter leur fonctionnement aux évolutions de leur environnement. Les résultats obtenus indiquent la pertinence d'utiliser de tels systèmes adaptatifs pour résoudre les problèmes inhérents au concept de Smart Grid. / Electricity plays an increasingly important role in our society. Indeed, we are moving toward the era of "everything electric". The needs evolving, it is mandatory to rethink the way electricity is produced and distributed. This then introduces the concept of an autonomous and intelligent power system called the Smart Grid. The Smart Grid is a concept of electrical network able to support autonomously any changes and faults that may occur. Obviously, the geographical distribution of electrical networks and the environment (weather conditions, ...) make it impossible to predict events that will occur. To do this, this study proposes an innovative agent-based framework as well as the design and implementation of cooperative agents behaviors aiming at solving common power systems related problems: the Load Flow analysis and the State Estimation. These issues have been addressed by the mean of Adaptive Multi-Agent Systems. These systems are known to be efficient to solve complex problems and have the ability to adapt their functioning to the evolutions of their environment. The results obtained show the relevance of using such self-adaptive systems to solve the issues inherent to the Smart Grid.

Cette thèse présente l'utilisation conjointe de la théorie de l'évidente et du suivi multi-hypothèses pour la modélisation et la gestion d'un système de suivi multi-caméras dans un environnement autoroutier. Le suivi est basé sur la ré-identification des objets (véhicules) sur la base d'information visio-temporelles. Une concrétisation de ces concepts se traduit par la conception et la mise en oeuvre d'une architecture logicielle multi-agents de gestion du suivi multi-caméras. Après une présentation de l'état de l'art sur les cadres de gestion de l'incertain et celui relatif à fusion de l'information pour la mise en correspondance, et sur les systèmes multi-agents, notre apport dans ce travail se situe à trois niveaux. Le premier a été une adaptation de la phase de décision du modèle de croyance transférable pour y intégrer l'utilisation du suivi multi-hypothèses comme outil de levée d'ambigüité rn cas d'indécision face à une situation de mise en correspondance. Le second apport a été celui de proposer une architecture logicielle à base d'agents pour la gestion du système du suivi multi-caméras. Nous en avons proposé la modélisation globale ainsi que celle des agents et de leurs interactions en utilisant une démarche personnelle d'analyse mais toutefois inspirée de langages et outils de modélisation tels que Agent UML et MaSE pour ne citer que ceux-là, du fait qu'il n'existe pas réellement un standard normalisé à ce jour dans ce domaine. Notre troisième apport a été de faire un début d'implémentation de notre architecture logicielle à base d'agent en nous basant sur la plateforme JADE (Java Agent DEvelopment Framework). Quelques expérimentations et discussions des résultats sont présentées à la fin pour déboucher sur nos conclusions et perspectives. / This thesis presents the joint use of the theory of evidence and multiple hypothesis tracking for modeling and managing a system for monitoring multiple cameras in a motorway. The tracking is based on the re-identification of objects (vehicles) on the basis of visuals and times informations. A realization of these concepts results in the design and implementation of a software architecture for multiple agents management of multiple camera tracking system. After presenting the state of the art on the frameworks of uncertainty management and that on information fusion for the matching, and the multi-agent systems, our contribution in this work is on two or three levels. The first was an adaptation of the decision phase of the transferable belief model to incorporate the use of multi-hypotheses tracking as a tool of ambiguity survey in case of indecision in matching situation. The second contribution was a proposition of agent-based software architecture for management of a multiple cameras tracking system. We have proposed the global system modeling as well as agents and their interactions modeling using a personal analysis method but nevertheless inspired by modelisation languages and tolls such as Agent UML, MaSE and others, because there is not yet a standard and normalized tool on the subject. Our third contribution was to begin an implementation of our agent-based software architecture using JADE (Java Agent Development Framework). Some experiment and discussions are presented at the end to lead to our conclusions and perspectives.

Les Agents Conversationnels Animés (ACA) sont des personnages virtuels interactifs et expressifs, dont l'aspect est très souvent « humain », exploitant différentes modalités telles que la face, le langage, les gestes, le regard ou encore la prosodie de la voix. Le but est qu'ils s'expriment en langage naturel et puissent dialoguer avec des interlocuteurs humains. Pour développer un ACA, il faut d'abord comprendre que des aspects tels que personnalité, les émotions et leur apparence sont extrêmement importants. Le travail qui est présenté dans cette thèse a pour objectif d'augmenter l'acceptabilité et la crédibilité des agents au moyen de la personnalité, considérée comme une notion centrale à l'interaction ACA-humain. On propose un modèle qui dote l'ACA de facettes de personnalité et de buts de communication « cachés » et qui module ainsi ses actions conversationnelles. Ce travail présente également une application de jeu de type “puzzle”, intégrant un ACA doté de facettes de personnalité et de buts « cachés », qui a servi de support à plusieurs expérimentations et à l'évaluation du modèle proposé. / Embodied Conversational Agents (ECAs) are intelligent software entities with an embodiment used to communicate with users, using natural language. Their purpose is to exhibit the same properties as humans in face-to-face conversation, including the ability to produce and respond to verbal and nonverbal communication. Researchers in the field of ECAs try to create agents that can be more natural, believable and easy to use. Designing an ECA requires understanding that manner, personality, emotion, and appearance are very important issues to be considered. In this thesis, we are interested in increasing believability of ECAs by placing personality at the heart of the human-agent verbal interaction. We propose a model relating personality facets and hidden communication goals that can influence ECA behaviors. Moreover, we apply our model in agents that interact in a puzzle game application. We develop five distinct personality oriented agents using an expressive communication language and a plan-based BDI approach for modeling and managing dialogue according to our proposed model. In summary, we present and test an innovative approach to model mental aspects of ECAs trying to increase their believability and to enhance human-agent affective communication. With this research, we improve the understanding on how ECAs with expressive and affective characteristics can establish and maintain long-term human-agent relationships.

Made available in DSpace on 2016-06-02T19:06:14Z (GMT). No. of bitstreams: 1 6183.pdf: 2878708 bytes, checksum: f9bc336337651cbba67af52d8acb7ec2 (MD5) Previous issue date: 2014-06-10 / Financiadora de Estudos e Projetos / Vehicle fleets are an important component in several applications, moving materials and people. Examples include material handling in warehouses, factories and port terminals, people transportation as in taxi fleets and emergency services, such as medical assistance, fire-fighters and police. Fleet operation is crucial for these applications: it can mean loss of money and commercial partners in case of industry, os loss of lives in case of emergency services. Controlling the fleet to achieve efficient levels of performance is a difficult problem, though, and becomes even harder as the fleet grows. Research in the area has been linking vehicle fleet operation to Multi-Agent Systems, because vehicle fleets are naturally distributed and Multi-agent System is a convenient abstraction to cope with distributed Artificial Intelligence problems. Therefore, it is proposed a Multi-Agent System to control vehicle fleets, focusing on material handling application in warehouses. The proposed system has three types of agents: Vehicle Agent, Loading Point Agent and Storage Point Agent. Agents interact amongst themselves through messages, trying to efficiently realize the material handling in a warehouse. System implementation is done through a simulation of a warehouse operation, built on top of MASON multi-agent system simulation platform. Task assignment strategies is also an important problem, therefore four strategies are shown and tested using the simulation: CNET, Fuzzy, DynCNET and FiTA. To enable comparison among these strategies, a Genetic Algorithm is employed to systematically search good parameters for each strategy. The proposed system, as well as the simulation, are offered as a framework for development of other vehicle fleets controlling multi-agent systems and/or task assignment strategies. / Em várias aplicações, frotas de veículos são um componente importante, transportando materiais e pessoas. Exemplos incluem o manejo de materiais em depósitos, fabricas e terminais portuários, o transporte de pessoas como em frotas de taxis e serviços de emergência, como socorro medico, bombeiros e polícia. A operacao da frota e crucial para essas aplicações: pode significar perda de dinheiro e parceiros comerciais no caso dos exemplos na indústria, ou perda de vidas, no caso de serviços de emergência. Porem, controlar a frota de modo que ela opere eficientemente e um problema difícil, que se torna ainda mais custoso com o aumento da frota. Pesquisas na área tem ligado a operação de frotas de veículos a Sistema Multiagente, notando os fatos de que frotas de veículos são naturalmente distribuídas e que o conceito de Agentes (e, consequentemente, Sistemas Multiagentes) e uma abstração conveniente para lidar com problemas de Inteligencia Artificial de forma distribuída. Com base nisto, e proposto um Sistema Multiagente para controle de frotas de veículos, focando a aplicação dessa frota no manejo de materiais em um depósito. O sistema proposto possui três tipos agentes: Agente de Veículo, Agente de Ponto de Carga e Agente de Ponto de Armazenamento. Os agentes interagem entre si, trocando mensagens a fim de realizar o manejo dos materiais no deposito de forma eficiente. O sistema e implementado na forma de uma simulação de operação de um deposito, construída na plataforma de simulação de sistemas multiagentes MASON. Como a estrategia de associação de tarefas também e um problema importante, quatro estratégias são mostradas e testadas através da simulação: CNET, Fuzzy, DynCNET e FiTA. Para possibilitar comparações entre as estrategias, um Algoritmo Genetico foi utilizado para sistematicamente encontrar bons parâmetros para as quatro estrategias. O sistema proposto, bem como a simulação, são oferecidos como framework para construção de outros sistemas multiagentes para frotas de veículos e/ou estrategias de associação de tarefas.

A capacidade de reconfiguração de sistemas de manufatura tem sido procurada pelas empresas para assegurar características de agilidade, eficiência e exibilidade para atender as mudanças de tipo/quantidade de produtos, processos, recursos e, além disso, para assegurar a devida reação à ocorrência de falhas. Por outro lado, a Indústria 4.0\" impõe novos desafios para os sistemas de controle, tais como a integração de tecnologias de interação entre homem e máquina em cadeias de valor compondo uma rede de plantas industriais geograficamente dispersas. O controle de sistemas reconfiguráveis de manufatura deve considerar: (i) funcionalidades de sistemas distribuídos e dispersos, tais como agilidade de resposta às mudanças, autonomia e colaboração entre os componentes para alcançar os objetivos do sistema de forma conjunta; (ii) interfaces para sua interoperabilidade e portabilidade; (iii) modularização para facilitar a manutenção, expansão e atualização do sistema, evitando a sobreposição de escopos; e (iv) mecanismos de controle para supervisionar as ações e interações dos componentes, o diagnóstico e a tomada de decisão. O pleno atendimento a estes requisitos não é trivial e formalismos para o desenvolvimento de soluções devem ser adotados. Uma solução é combinar técnicas voltadas para sistema multiagente e holon com arquitetura orientada a serviço através de uma adequada técnica de modelagem usando extensões de rede de Petri: Production Flow Schema e Input Output Place Transition. Portanto, este trabalho prop~oe uma arquitetura de controle e o método de modelagem de seus componentes para sistemas reconfiguráveis de manufatura combinando estas técnicas e considerando os aspectos de personalização, convertibilidade, escalabilidade, modularidade, integrabilidade, diagnosticabilidade, interoperabilidade e colaboração entre os componentes do sistema de controle, inclusive do homem. Um exemplo de aplicação é apresentado para demonstrar a viabilidade da proposta e comprovar os resultados alcançados. / The reconfiguration ability of the manufacturing systems has been approached by companies to ensure agility, efficiency and exibility characteristics to address the changes of type/quantity of products, processes and resources and, furthermore, to ensure proper reaction to the fault occurrence. On the other hand, the\"Industry 4.0\" imposes new challenges for control systems, such as interaction between man and machine into value chains composing a network of geographically dispersed industrial plants. The control of reconfigurable manufacturing systems should consider: (i) functionalities requirements of distributed and disperse systems, such as responsiveness to changes, autonomy and collaboration among components to achieve the global system aim; (ii) interfaces for interoperability and portability; (iii) modularity to facilitate maintenance, expansion and upgrade of the system, avoiding the overlapping of scopes; and (iv) control mechanisms to supervise the actions and interactions among components, diagnosis and decision making. The compliance with these requirements is not trivial and formalisms to develop solutions must be adopted. A solution is combining techniques based on holonic and multi-agent system with service-oriented architecture through appropriate modeling using Petri net extensions: Production Flow Schema and Input Output Place Transition. Therefore, this paper proposes control architecture and a method to model components for reconfigurable manufacturing systems, combining these techniques and considering aspects of customization, convertibility, scalability, modularity, integrability, diagnosability, interoperability and collaboration among control system components, including humans. An application example is presented to demonstrate the feasibility of the proposal and verify the results.

Submitted by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2016-02-04T17:56:58Z No. of bitstreams: 1 Dissertação - Neila Batista Xavier.pdf: 16096870 bytes, checksum: 14fa0cdac4b0ae0046b9b6a6b2e06c88 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2016-02-04T17:57:48Z (GMT) No. of bitstreams: 1 Dissertação - Neila Batista Xavier.pdf: 16096870 bytes, checksum: 14fa0cdac4b0ae0046b9b6a6b2e06c88 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2016-02-04T17:58:07Z (GMT) No. of bitstreams: 1 Dissertação - Neila Batista Xavier.pdf: 16096870 bytes, checksum: 14fa0cdac4b0ae0046b9b6a6b2e06c88 (MD5) / Made available in DSpace on 2016-02-04T17:58:07Z (GMT). No. of bitstreams: 1 Dissertação - Neila Batista Xavier.pdf: 16096870 bytes, checksum: 14fa0cdac4b0ae0046b9b6a6b2e06c88 (MD5) Previous issue date: 2015-10-09 / With the continued and increasing use of Learning Management System (LMS) in educational institutions, it is necessary for coordinators and administrators effectively manage the online courses in order to ensure good monitoring and superior levels looking for ways to improve the management of online courses. Thus, considering the difficulties in the aspect of infrastructure and time to become familiar with the resources available, the researchers realized the need to know the opinion of people with experience in Distance Education and higher level positions to decision-making. During the investigation of the problem, was held collection and analysis of information on the performance of LMS in the follow-up courses process. The results revealed the need to implement a tool that would help the monitoring of online courses. Based on this information, this paper proposes and describes an approach based on a Multi-Agent System for this problem. The system has the purpose to support the management of online courses through intelligent agents. From the use of the proposed system, the coordinator of an online courses get condensed and quick information of each current course, facilitating managerial vision by reporting and organized graphics, and receive periodically a message via email about the current status of each course. Thus, intelligent agents show specific situations of access students, tutors and teachers, and help the follow-up activities. To validate the work, was carried out a case study in the postgraduate courses in online mode with several real disciplines concurrently taking place in a Federal Institute of Education (IFE) of Amazonas state. The case study aimed evaluate the feasibility of the proposal, and from the results obtained in the test can be concluded that with using the developed system was possible to improve the monitoring of levels of online courses, assist in making decisions and thus enable the adoption of new teaching strategies / Com o contínuo e crescente uso de Ambientes Virtuais de Aprendizagem (AVA) em instituições de ensino, é necessário que coordenadores e administradores gerenciem efetivamente os cursos de Educação a Distância (EaD), a fim de garantir um bom acompanhamento e níveis de qualidade superior, buscando meios para aperfeiçoar o gerenciamento de cursos on-line. Assim, considerando as dificuldades no aspecto de infraestrutura e tempo para familiarizar-se com os recursos disponíveis, os pesquisadores perceberam a necessidade de conhecer a opinião de pessoas com experiência em EaD e em cargos de nível hierárquico superior para tomada de decisões. Durante a investigação do problema, foi realizada a coleta e a análise de informações sobre o desempenho de AVAs no processo de acompanhamento de cursos. Os resultados revelaram a necessidade da implantação de uma ferramenta que auxiliasse o acompanhamento dos cursos online. Com base nessas informações, o presente trabalho propõe e descreve uma abordagem baseada em um Sistema Multiagente (SMA) para esse problema. O sistema tem como proposta apoiar a gestão de cursos on-line por meio de Agentes Inteligentes (AIs). A partir do uso do sistema proposto, o coordenador dos cursos de um AVA obtêm informações condensadas e rápidas de cada curso corrente, facilitando a visão gerencial mediante a apresentação de relatórios e gráficos organizados, além de receber, periodicamente, uma mensagem via e-mail a respeito da situação atual de cada curso. Desta forma, os AIs evidenciam situações específicas do acesso de alunos, tutores e professores, além de auxiliar o acompanhamento de atividades. Para validar o trabalho, foi realizado um Estudo de Caso em cursos de Pós-Graduação na modalidade EaD com várias disciplinas reais ocorrendo concomitantemente em um Instituto Federal de Educação (IFE) do estado do Amazonas. O Estudo de Caso teve como objetivo avaliar a viabilidade da proposta, e a partir dos resultados obtidos nos testes realizados pode-se concluir que com o uso do sistema desenvolvido foi possível melhorar os níveis de acompanhamento dos cursos on-line, auxiliar no processo de tomada de decisões e, assim, possibilitar a adoção de novas estratégias pedagógicas.

Made available in DSpace on 2015-04-11T14:02:58Z (GMT). No. of bitstreams: 1 Andre Alves Nogueira.pdf: 3385945 bytes, checksum: 5c7b555b59610b2dac767f2900bfe5be (MD5) Previous issue date: 2014-02-20 / Weather Extreme Events (WEE) can cause natural disasters because they are often accompanied by dangerous physical activities which cause significant damage to people and properties, including some deaths. A way to soften the damage caused by these disasters is to use Weather Warning System (WWS). These kind of systems can detect an imminent risk and, different types of alerts can be triggered in order to mobilize the emergency personnel in advance. This research aims to help the Amazon Protection System (SIPAM) to solve the communication problem with the alert and mobilization. It is intended, primarily, to assist the Civil Defense with the affected population by WEE. A comparative investigation was performed in ten studies aiming to understand and analyze the main characteristics of a Warning System, which were incorporated into this research, highlighting the support of a Multi-Agent System to assist the WWS, and the use of SMS (Short Message Service) technology and E-mail for sending the alert. System Tests showed that the approach proved promising, with shipping times smaller alerts those found in the literature with a high rate of confirmed users. Also the process of sending the alert factor being faster helps to minimize the effects of EME with the affected population. / Eventos Meteorológicos Extremos (EME) podem gerar desastres naturais, geralmente acompanhados de atividades físicas perigosas que provocam danos significativos a pessoas e propriedades, gerando um grande número de vítimas ou até mesmo mortes. Uma maneira de mitigar os danos causados por estes desastres é a utilização de Sistemas de Alertas Meteorológicos (SAM); sendo detectado um risco iminente, diferentes tipos de alertas podem ser acionados objetivando a mobilização antecipada das equipes de emergência. O presente trabalho visa contribuir com o Sistema de Proteção da Amazônia (SIPAM) ou com órgãos semelhantes para a resolução do problema de comunicação referente à emissão dos alertas e a mobilização, destinando-se principalmente à ação da Defesa Civil e órgãos competentes junto à população em função dos EME ocorridos. Nesta pesquisa foi realizada uma investigação comparativa com dez trabalhos objetivando conhecer e analisar as principais características de um Sistema de Alerta, as quais foram incorporadas neste projeto, destacando-se o apoio de um Sistema Multiagente para auxiliar o SAM, e o uso das tecnologias de SMS (Short Message Service) e E-mail para o envio do alerta. Testes do sistema evidenciaram que a abordagem adotada mostrou-se promissora, apresentando tempos de envio de alertas menores que os encontrados na literatura e com alta taxa de usuários confirmados, fator de muita importância para a minimização dos efeitos dos EME junto à população atingida.

Made available in DSpace on 2016-03-15T19:37:56Z (GMT). No. of bitstreams: 1 KHAOHUN SHIA.pdf: 2242262 bytes, checksum: 97a46586e1ac55fd1097f77a932ad878 (MD5) Previous issue date: 2015-08-25 / In a competitive market of software, companies seek an agile development process with quality. Faced with this challenge, the companies target is a certification to prove the ability to produce a high quality product by following a defined process, for example, Capability Maturity Model Integration for Development (CMMI-DEV). To achieve this goal, the company hires experts to change the existing process, which represents a high cost for small and medium-sized businesses. The assumption in this work is to evaluate the possibility of using an agile methodology which adheres to CMMI-DEV. Feature Driven Development (FDD) was the agile methodology selected to assess their adherence to Project Planning area of CMMI-DEV. The evaluation of specific practices was automated by the multi-agent system with two ontologies. The first of them is the Ontology CMMI/FDD, which defines the concepts and the relationship of FDD with these specific practices. The second is the Ontology of Project Planning Document that defines the concepts used in sections of the Project Planning documents. The mapping between the specific practices and implementation was performed by intelligent mapping agents. The results obtained shown that it is possible to automate the process of evaluation of specific practices by the evidences in sections of the documents, reducing the costs involved in hiring a specialist of CMMI. The evaluation of Intelligent Agents of Specific Practices shown that the FDD partially attends to the specific practices, being necessary to change the two documents elaborated from the FDD Document (Project Plan and List of Functionality) and add three auxiliary documents (Risk Management, Data Management and Training Management) to meet all the specific practices of the Project Planning Area of the CMMI-DEV. / Em um mercado competitivo de desenvolvimento de software, as empresas buscam um processo ágil com qualidade. Diante desse desafio, procuram uma certificação para comprovar a capacidade de produzir com qualidade, seguindo um processo definido, por exemplo, o Capability Maturity Model Integration for Development (CMMI-DEV). Para atingir essa meta, a empresa contrata especialistas para alterar o processo existente, o que representa um alto custo para empresas de pequeno e médio porte. A hipótese neste trabalho é avaliar a possibilidade de utilizar uma metodologia ágil aderente ao CMMI-DEV. O Desenvolvimento Guiado por Funcionalidades (FDD) é a metodologia ágil selecionada para computar a sua aderência em relação à área de Planejamento de Projetos do CMMI-DEV. A avaliação das práticas específicas foi automatizada por um sistema multiagentes com duas ontologias. A primeira delas é a ontologia CMMI/FDD, que define os conceitos e a relação do FDD com essas práticas específicas. A segunda é a ontologia dos Documentos do Planejamento de Projeto, que define os conceitos utilizados nas seções dos documentos do Planejamento de Projeto. O mapeamento entre a prática específica e a implementação nas seções de documentação foi realizado pelos Agentes Inteligentes de Mapeamento. Os resultados obtidos mostram que é possível automatizar o processo de avaliação das práticas específicas, mostrando as evidências nas seções dos documentos, reduzindo os custos envolvidos na contratação de um especialista em CMMI. A avaliação dos Agentes Inteligentes de Práticas Específicas mostra que o FDD atende parcialmente às práticas específicas, sendo necessário alterar os dois documentos elaborados a partir do documento do FDD (Plano de Projeto e Lista de Funcionalidade) e adicionar três documentos auxiliares (Gestão de Risco, Gestão de Dados e Gestão de Treinamento) para satisfazer a todas as práticas específicas da área de Planejamento de Projeto.

La gestion de crise est devenue une activité indispensable à toute organisation publique ou privée. Elle repose le plus souvent sur un dispositif particulier appelé « cellule de crise » qui a pour objectif de mettre en œuvre des mesures d'anticipation, de vigilance, d'intervention et de sauvegarde et fait intervenir de multiples acteurs qui interagissent. Pour se préparer aux situations de crise, les communes ont l'obligation de définir un PCS (Plan Communal de Sauvegarde), mais ce dernier ne suffit pas à lui seul à garantir l'efficacité de l'organisation de crise. L'objectif de ce travail est de proposer une méthodologie, principalement fondée sur la pensée systémique et sur les Systèmes Multi-Agents, pour étudier la vulnérabilité organisationnelle de la cellule de crise et en évaluer sa performance. Un outil de simulation a été développé à destination des responsables. Il intègre des indicateurs de performance organisationnels pour identifier les défaillances potentielles de l'organisation. Les résultats de la simulation sont consignés sous la forme d'un tableau de bord dont l'analyse peut servir à améliorer l'organisation de crise. / Crisis management has become an essential activity in public and private organizations. It is most often supported by a "crisis cell" that aims to implement anticipation, vigilance, response and backup measures by involving multiple actors interactions. To be prepared for emergencies, municipalities have to define a EOP (Emergency Operations Plan). However, EOP on its own is not sufficient to ensure the effectiveness of crisis organization. The objective of this work is to propose a methodology mainly based on systems thinking and Multi-Agent Systems, to study the organizational vulnerability of crisis cell and assess its performance. A simulation tool is developed for decision makers. Organizational performance indicators are integrated to identify potential failures of the organization. Simulation results are reported in the form of a dashboard whose analysis can be used to improve crisis organization.

Context : Alzheimer's disease is a memory impairment disease which mostly affects elderly people. Currently, about 4 million Americans and 5 million Europeans are affected by this disease. The occurrence of Alzheimer's disease is expected to quadruple by the year 2020. Alzheimer's disease cannot be cured or stopped its progression rather delay its progression. Early diagnosis of the disease helps the patients, the caregivers and health institutions to save time, cost and minimize patients suffering. Objectives : In this thesis, different machine learning algorithms used for classification purpose are evaluated and various Alzheimer's disease diagnosis techniques are identified. Among these algorithms, a suitable classifier that has better classification accuracy on the National Alzheimer Coordinating Center (NACC) dataset is selected. This classifier is customized in order to make it compatible for the NACC dataset and to receive the new instance from the user. Then a multi-agent system model is develop that can improve the classification accuracy. Methods : Different research works are reviewed and experiments are conducted throughout this research work. A dataset for this research is obtained from National Alzheimer's Coordinating Center, university of Washington. Using this dataset, two experiments are conducted in WEKA. In the first experiment, the five candidate algorithms are compared to select the significant classifier for medical history and cognitive function data. For the second experiment, two datasets are used; a dataset contains Medical History (MH) with Cognitive Function (CF) data and a dataset that contains only medical history data to check in which dataset the selected classifier has better accuracy. Results : From the first experiment, J48 classifier has a better stage prediction accuracy than the candidate algorithms with 61.12%. J48 is customized to classify a new instance received from the user and to improve the classification accuracy. Then the accuracy increase to 87.09% when the classifier's parameters are optimized. When the medical history and cognitive function data is experimented in WEKA separately, the classification accuracies of J48 on MH, CF and their combination datasets are 81.42%, 64.20% and 87.09% respectively. The agents simulation result showed that some misclassified instances by J48 algorithm can be corrected by using multi agent system. The experimental results are presented in graphical format. Conclusions : Hence we conclude that machine learning and agent system in combination can be used for Alzheimer's disease diagnosis and its stage prediction by extracting knowledge from a dataset which contains patients medical history and cognitive function data. / Syftet med detta examensarbeta var att diagnostiserar Alzheimer patienter använder mönstret från en samling av andra tidigare diagnoserad patienter information och diagnosdata. Examensarbete hade tre huvuduppgifter: Förberedelse av data (mer än 10000 patienter data) för forskningen, maskininlärning algoritmer utvärderade med WEKA verktyg för att välja den bästa algoritmen och förbättra noggrannheten av den valda algoritmen med hjälp av agent system tekniker . - SQL queries används på uppgifter förberedelsefas. - WEKA programvara används för algoritmer utvärdering. - Agent arkitektur är utvecklat för att förbättra förutsäga av noggrannhets. Det bidrag av detta examensarbeta är identifiera Alzheimer patienter diagnos metod som använder en samling av patienternas diagnos information / biliyala.ezd2@gmail.com, them22dayz@gmail.com