Route optimization aims to optimize routes for vehicles withregards to resource usage. Especially when the vehicle needsto visit multiple customers on the route, a route optimizationtool is beneficiary. The purpose of this study is to comparehuman experience with a route optimization program. This isdone by comparing how a truck driver makes his routes to theroute a GIS-tool has calculated and then see which of theroutes was shorter, measured in kilometers. The data for thisstudy was gathered from a big shipping company. In order toachieve the purpose of this study 10 routes were analysed bya GIS program called ArcGIS. The algorithm used by ArcGISin route optimization is tabu search, this type of program wasused because it is based on heuristic methods that is muchfaster than exact methods. Expert systems are based onknowledge from experts that have been accumulated duringmany years of experience. Providing recommendations basedon probability reasoning instead of absolute answer. Thesekind of systems is often used in GIS programs to improveresults and calculation time. The aim of this study wasanalyze if a optimization program finds a better route than theexpert. This study shows an improvement of 60% of theanalyzed routes. To verify the results of this study anhypothesis test was made which gave a level of significanceby more than 85 %. The routes were optimized to a certainextent even before the study was done due to the driveralready being familiar with the routes in question. Because ofthis the results of this study were lower compared to othersimilar studies. Another reason may be that the coordinatesgiven to us did not always correspond perfectly with actuallocation of the stops. / Ruttoptimering avser att optimera rutter för fordon medminsta möjliga resursåtgång. När fordonet ska besöka ettflertal givna platser är ett ruttoptimeringsverktyg förmånligtatt använda. Denna studie syftar till att jämföra den mänskligaerfarenheten mot ett ruttoptimeringsprogram. Detta har gjortsgenom att jämföra hur en lastbilschaufför har kört en rutt mothur ett GIS-verktyg räknat fram den optimerade färdvägen avsamma rutt. Sedan jämfördes om det fanns skillnader ochvilken av rutterna som var kortast, räknat i kilometer. Datahar hämtats från ett stort fraktföretag. För att nå syftet har 10rutter undersökts i programmet ArcGIS Online som använderalgoritmen tabusökning. En kommersiell beräkningsmetodhar använts då det bygger på heuristiska metoder som ärbetydligt snabbare än exakta metoder. Expertsystem byggerpå erfarenhet som experter har samlat på sig genom åren, deger rekommendationer baserade på sannolikhetsresonemangistället för definitiva svar, dessa system sätts ofta in i GIS för att förbättra resultat och beräkningstider i systemen. Studienresulterade i en förbättring på 60 % av rutterna. Målet meddenna undersökning var att visa om ett optimeringsprogramhittar en bättre rutt än experten. För att verifiera resultaten istudien gjordes en hypotesprövning vilket gav ensignifikansnivå på över 85%. Chauffören har kört dessa rutteri flera år vilket gör att rutterna är optimerade i en viss månredan innan studien gjordes. Det har inverkat på resultatetsom gett ett lågt medelvärde av den procentuella skillnaden,jämfört med tidigare undersökningar. En annan faktor kanvara att koordinaterna i datan från företaget inte helt stämdemed den verkliga placeringen av stoppen på rutterna.

Dans cette thèse, nous fournissons un examen complet des technologies du Web sémantique et de leurs utilités dans le contexte actuel des petites et moyennes entreprises (PME). Les approches traditionnelles d'intégration des entreprises favorisent essentiellement les grandes entités. Les obligations contractuelles fortes sur les PME, mais en même temps leur volonté de garder leurs compétences individuelles, et ce, dans un environnement limitant leur choix, les obligent à prendre des décisions stratégiques et de conclure des accords sur le long terme avec leurs partenaires, limitant ainsi leur flexibilité aux fluctuations du marché. Nous proposons, donc, une approche ontologique basée sur Web sémantique pour l'intégration de l'information ainsi que des ressources matérielles de l'entreprise distribuée. Cette approche, basée sur le Web, agit comme un système d'aide à la décision pour utiliser des ressources de meilleure qualité ainsi que pour l'intégration de l'information distribuée. Les travaux relatifs à l'ontologie web, pour l'intégration d'information ne sont pas nouveaux, mais l'approche proposée par nous est une valeur ajoutée pour l'entreprise distribuée. De plus, nous avons également proposé l'ontologie Web sémantique comme un système de configuration pour gérer les ressources distribuées de l'entreprise virtuelle. Puis, nous avons modélisé l'ontologie OWL-DL en nous basant sur la sémantique de la norme ISA-95, relative à l'intégration d'entreprises industrielles. Ensuite, nous utilisons cet artefact ontologique comme un artefact de configuration permettant de gérer le matériel de l'entreprise virtuelle distribuée ainsi que les ressources matérielles. C'est la proposition principale de cette thèse : utiliser l'ontologie Web sémantique comme un système d'aide à la décision pour la configuration de l'utilisation des ressources / In this thesis, we have provided a complete review of the semantic web technologies and their corresponding utility in the current environment for small to medium sized enterprise (MSE). The traditional approaches to enterprise integration favour large enterprise entities and force contractual limitations on smaller partners, but at the same time the pressure to guard the individual enterprise competence is ever increasing, the distributed enterprise (MSE) in such an environment have limited number of choices, which forces them to make strategic decisions and enter into a long term agreements with their partners and this limits their flexibility to the market changes. We, in this thesis, propose a semantic web based ontology approach for integrating the information as well as physical resource of the distributed enterprise. This web based approach acts as a decision support for better resources utility as well as distributed information integration. The work related to web ontology?s for information integration is not new, but the approach proposed in this thesis for distributed enterprise is an added value. Similarly, we have also proposed semantic web ontology as a configuration system to manage the distributed resources of the virtual enterprise, for this we have modelled OWL-DL ontology on the semantic of the industrial integration standard ISA-95, and subsequently used this ontology artefact as a configuration artefact to manage the distributed virtual enterprise material and equipment resources this is the main proposition of the thesis of utilizing semantic web ontology as resource configuration decision support

Es wurde ein Expertensystem zur Prognoseabschätzung entwickelt und evaluiert, welches eine neue Art der Regelbewertung anwendet. Als Krankheitsbild wurden exemplarisch Hirnstammgliome im Kindesalter gewählt. HISTAGLI besteht aus einer Informationskomponente, die einen inhaltlichen Überblick über das Krankheitsbild gibt, einer Datenbank, in der sämtliche Patienteneingaben sowohl als Text, als auch graphisch einsehbar ist und einer Prognoseerstellungskomponente, in der eine Prognose in sechs verschiedenen Kategorien unter Berücksichtigung verschiedener Therapieschemata für einen neu eingegebenen Patienten erstellt und der Weg zur Prognoseerstellung erläutert wird. Die der Prognoseerstellungskomponete zugrundeliegende Wissensbasis wurden aufgrund der klinischen und histopathologischen Daten von 23 Kindern, die an einem Hirnstammgliom erkrankt waren, halbautomatisch von einem Experten bewertet. Es fand sich dabei eine hohe Korrelation von 78,26% (bzw. 86,96% bei einer Kategorie Toleranz) Übereinstimmung mit den tatsächlichen Prognosen der Patienten. Dieses Ergebnis ist nur mit Einschränkungen auf neue Patienten übertragbar, da die Fallzahl zu gering für statistische Aussagen ist. Bei entsprechender Pflege des Systems ist mit wachsenden Fallzahlen mit einer immer höher werdenden Genauigkeit zu rechnen. / An expert system for the estimation of a prognosis was developed and evaluated which uses a new kind of valuation. Brainstem gliomas in the childhood were chosen as an example of a disease. HISTAGLI consists of an information component which gives an overview about the disease, a database in which all patient data is presented as text and graphics, and a prognosis creation component which creates a prognosis for newly inserted patients in six categories under consideration of different therapy pattern and explains the way towards the estimation of the prognosis. The knowledge base of the prognosis creation component was made out of the clinical and histopathological data of 23 children with brainstem gliomas halfautomaticly valued by an expert. There was a high correlation of 78,26% (with one category toleranz 86,96% ) accordance with the real prognosis of the patients. This result is only restricted transferable to new patients because of the number of cases which is too small for statistical evidence. Higher precision is expected with appropriate maintenance and an increasing number of cases.

Esta tese assumiu a premissa de que a neuropatia periférica é um sinal de piora da diabetes, além de levantar a questão de que estudos prévios sobre a biomecânica da marcha de diabéticos não têm distinguido os graus de progressão da diabetes nos grupos estudados. Neste contexto, não é possível identificar as diferenças nos padrões de geração da marcha entre estágios precoces e avançados da diabetes. Esta identificação poderia facilitar a intervenção terapêutica precoce nestes pacientes, o que poderá impedir a formação de úlceras e amputações recorrentes subseqüentes. Assim, apresentamos ao longo desta tese, três estudos para investigar a natureza das supostas alterações na marcha (estudo 1) e no descer escadas (estudo 2) de diabéticos, assim como para propor uma forma de classificar a progressão da diabetes levando em consideração as incertezas de fronteiras entre os subgrupos de neuropatas, por meio de um sistema especialista fuzzy (estudo 3). Os estudos 1 e 2 foram feitos com os mesmos três grupos: indivíduos diabéticos (GD) e diabéticos neuropatas (GDN) diagnosticados clinicamente e indivíduos saudáveis (GC). Para a avaliação cinemática e cinética do membro inferior foram utilizadas câmeras infravermelhas e uma plataforma de força durante o andar no plano e descendo uma escada. O cálculo dos momentos articulares de membro inferior foi feito por meio do método da dinâmica inversa. Os principais resultados do estudo 1 mostraram que independente da presença da neuropatia, os pacientes diabéticos exibiram uma maior flexão das três principais articulações do membro inferior e um importante uso da articulação do quadril como uma estratégia cinética de progressão do corpo à frente, em substituição ao tornozelo, que mostrou ser a articulação mais prejudicada. Os principais resultados do estudo 2 indicaram as mesmas mudanças significativas no padrão cinemático do tornozelo durante a fase de propulsão, mesmo na ausência da neuropatia. No entanto, não houve diferença nos padrões cinéticos entre os estágios iniciais e avançados da doença, mas mostraram a mesma tendência observada no estudo 1 de alteração do padrão cinético de quadril para se adaptar às perdas distais nos neuropatas. No estudo 3, desenvolvemos um modelo para classificação da severidade da neuropatia diabética. O modelo fuzzy apresentou um nível de concordância adequado com a classificação feita por especialistas, e também mostrou uma alta precisão na avaliação de pacientes reais que foram submetidos à avaliação do modelo. O modelo foi capaz de simplificar e separar os pacientes em quatro diferentes graus de severidade, o que pode melhorar a eficácia de medidas preventivas, bem como para oferecer uma melhor ajuda para os profissionais de saúde no tratamento destas doenças e suas complicações. Como conclusão geral temos que os grupos diabéticos estudados exibiram comportamentos biomecânicos durante o andar no plano e descendo degraus que são muitas vezes parecidos entre si (GD e GDN), poucas vezes diferentes entre si, mas na maioria das vezes bem distintos do grupo não diabético (GC), indicando que a questão da progressão não se esclareceu completamente ao separamos de maneira crisp os grupos em diabético e diabético neuropata. As perdas sensitivo-motoras-autonômicas impostas pela diabetes não podem ser subestimadas, uma vez que parecem ter início ainda quando a neuropatia não foi diagnosticada. Uma correta classificação do paciente pode antecipar a detecção dos níveis menos severos da doença, evitando esperar que os pacientes apresentem perdas irreversívies para inicar uma intervenção clínica eficaz e preventiva para preservar a locomoção independente destes pacientes / This study assumed the premise that the peripheral neuropathy is a sign of worsening of diabetes, as well as raising the issue that previous studies on the biomechanics of gait in diabetics do not have distinguished the degree of progression of diabetes in both groups. Therefore, it is not possible to identify differences in patterns of gait generation between early and advanced stages of diabetes. This identification would facilitate early therapeutic intervention in these patients, which could prevent the formation of recurrent ulcers and subsequent amputations. We present throughout this thesis, three studies to investigate the nature of the alleged changes in gait (study 1) and stair descent (study 2) of diabetics, and to propose a way to classify the progression of diabetes, taking into account the uncertainties of boundaries between the subgroups of neuropathy through a fuzzy expert system (study 3). Studies 1 and 2 were performed with the same three groups: diabetics (GD) and diabetic neuropathic (GDN) diagnosed clinically and healthy subjects (CG). For the kinematic and kinetic evaluation of the lower limb, we used infrared cameras and a force plate during walking on a level walkway (10 m) and descent a staircase. The calculation of net joint moments of the lower limb was performed using the method of inverse dynamic. The main results of Study 1 showed that, regardless of the presence of neuropathy, diabetic patients exhibited a greater flexion of the three major joints of the lower limb and an important use of the hip joint as a kinetic strategy of progression the body forward, replacing the ankle, which proved to be the most affected joint. The main results of Study 2 showed the same significant changes in the pattern of the ankle kinematics during propulsion phase, even in the absence of neuropathy. However, there was no difference between the kinetic patterns in early and advanced stages of the disease, but showed the same trend observed in the Study 1: a change in the kinetic pattern of the hip to adapt de locomotion to distal loss in neuropathic subjects. In Study 3, the model developed for classification of severity of diabetic neuropathy showed an adequate level of agreement with the classification of experts, and also showed a high accuracy in the evaluation of real patients who underwent to the evaluation of the model. The fuzzy model was able to simplify and separate the patients into four different degrees of severity, which can improve the effectiveness of preventive strategies as well as to offer a better assistance to health professionals in the management of this disease. It is concluded that the studied diabetic groups exhibited biomechanical behavior during walking and descend stairs that are often similar to each other (GD and GDN), a few times different from each other, but most often very different from the non-diabetic group (GC), indicating that the issue of progression was not fully understood separating the groups in a crisp way as diabetic and diabetic neuropathic subjects. The sensory, motor and autonomic losses imposed by the diabetes can not be underestimated, since they seem to appear when the neuropathy is not diagnosed yet. A correct classification of the patient can anticipate the detection of less severe levels of the disease and avoid that the patients show an irreversible loss to start an effective intervention and preventive strategies to keep the independent locomotion of these patients

"A Single Function Agent (SiFA) is a software agent, with only one function, one point of view, and one target object on which to act. For example, an agent might be a critic (function) of material (target) from the point of view of cost. This research investigates the possibilities and implications of the SiFA concept, and analyzes the definition language, negotiation language and negotiation strategies of the agents. After defining a domain-independent set of agent types we investigated negotiation, analyzing which pairs/groups of agents have reason to communicate, and what the information passed between them should be, as well as what knowledge was needed to support the negotiation. A library for the CLIPS expert system shell was built, which allows development of SiFA based expert systems from domain independent templates. We will present two such systems, one as implemented for the domain of ceramic component material selection and the other (in development) for simple sailboat design. The effect of negotiation on the design process and the results are discussed, as well as directions for future research into SiFAs."

Submitted by William Justo Figueiro (williamjf) on 2015-07-18T13:12:39Z No. of bitstreams: 1 49c.pdf: 7911969 bytes, checksum: 0676e3d66dd0b95277c5a1fef4af7376 (MD5) / Made available in DSpace on 2015-07-18T13:12:39Z (GMT). No. of bitstreams: 1 49c.pdf: 7911969 bytes, checksum: 0676e3d66dd0b95277c5a1fef4af7376 (MD5) Previous issue date: 2011 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O constante aumento no volume de informações em diversas áreas do conhecimento, via produção de material bibliográfico, multimídia e web, fruto dos avanços de pesquisas e estudos científicos, tem gerado um problema que é o gerenciamento e integração da informação de diferentes fontes. Com isso, a busca desses conhecimentos acaba tornando-se complexa e custosa, visto o volume de informações em alguns domínios de conhecimento. No entanto, esse conhecimento é de grande importância aos especialistas para raciocinar e chegar a conclusões que são utilizadas em tomadas de decisão ou como meio de solucionar problemas do domínio. Com base nessas necessidades, essa dissertação propõe um sistema de auxílio à classificação que tem como meta, oferecer funcionalidades que permitam aos usuários especialistas realizar a manutenção de informações de um domínio na base de conhecimento e a disseminação dela entre os usuários do sistema, por meio de consultas e/ou sugestões acerca do conhecimento resultante. Onde os resultados apresentados pelo sistema, gerados por meio de um motor de inferência, são baseados em Sistemas Especialistas, que busca construir raciocínios a partir de informações que o usuário possui sobre as observações do cenário do domínio. Juntamente com o sistema de inferência é proposto a associação de imagens relacionadas ao conhecimento, como forma de ilustrar as informações e descrições, e algoritmos de Processamento de Imagens para a redução da subjetividade nos casos de dúvida em relação às características visuais do estudo de caso, visto a falta de definições discretas de algumas informações que descrevem o conhecimento. Como estudo de caso para demonstrar a aplicabilidade do sistema proposto, o trabalho foi focado no domínio da paleontologia, mais especificamente na classificação de espécies de Ostracodes, organismos que representam grande importância para a identificação de fontes petrolíferas. Como resultado do trabalho obteve-se um sistema robusto e genérico, permitindo o armazenando de um grande volume de informações, separado pela área do conhecimento e subdividido por domínios de trabalho. Avaliações realizadas com usuários demonstram a efetividade da ferramenta e apontaram para a evolução de funcionalidades. / The constant increase in the volume of information of several areas of knowledge, through the production of bibliographic, multimedia and web material, is a result of advancements of research and scientific studies, and has led to issues regarding the management and integration of it. Therefore, the search for this knowledge becomes complex and difficult, considering the amount of information in some domains. However, this knowledge shows great importance to experts, who can analyze this information and use it to reach conclusions that are used in decision making or as a way for solving specific problems. Based on these requirements, this dissertation proposes a system to aid classification task, aiming to provide functionalities that allow the expert to maintain domain information in a knowledge base. Also, it is possible to disseminate this information to the others users through searches and/or suggestions, presenting information about the resulting knowledge. The results presented by the system are generated by an inference engine based on Expert Systems, which seeks to produce reasoning from users information about the domain. Along with the inference system is proposed a combination of images related to the knowledge as a way of illustrating the information and descriptions. Also, Image Processing algorithms are employed to reduce subjectivity in cases of uncertainty, regarding the visual characteristics of the case of study, considering the lack of discrete definitions for some information that describe the knowledge. A system was proposed for integrating the Expert Systems and Image Processing techniques. As a way to demonstrate the system applicability, a pertinent problem of paleontology domain focused in classification of Ostracodes species, which have great importance to petroleum exploration, was developed. The system was validated with users and experts that highlight its contribution: able to concentrate and store a big volume of information of many domains, incorporating support decision by image processing, and to be precise to map the expert knowledge. The experts cited as a major contribution that the system represents an application destined to inexperienced users, like students and novice researchers, used in the learning/training process.

La reconnaissance des perturbations survenant sur les réseaux HTA est une problématique essentielle pour les clients industriels comme pour le gestionnaire du réseau. Ces travaux de thèse ont permis de développer un système d’identification automatique. Il s’appuie sur des méthodes de segmentation qui décomposent de manière précise et efficace les régimes transitoires et permanents des perturbations. Elles utilisent des filtres de types Kalman linéaire ou anti-harmoniques pour extraire les régimes transitoires. La prise en compte des variations harmoniques et de la présence de transitoires proches se fait à l’aide de seuils adaptatifs. Des méthodes de correction du retard a posteriori permettent d’améliorer la précision de la décomposition. Des indicateurs adaptés à la dynamique des régimes de fonctionnement analysés sont utilisés pour caractériser les perturbations. Peu sensibles aux erreurs de segmentation et aux perturbations harmoniques, ils permettent une description fiable des phases des perturbations. Deux types de systèmes de décision ont également été étudiés : des systèmes experts et des classifieurs SVM. Ces systèmes ont été mis au point à partir d’une large base de perturbations simulées. Leurs performances ont été évaluées sur une base de perturbations réelles : ils déterminent efficacement le type et la direction des perturbations observées (taux de reconnaissance moyen > 98%). / The recognition of disturbances affecting MV networks is essential to industrials and distribution system operators. The aim of this thesis work is to design a near real-time automatic system able to detect and identify disturbances from their waveforms. Segmentation methods split the disturbed waveforms into transient and steady-state intervals. They use Kalman filters or anti-harmonic filters to extract the transient intervals. Adaptive thresholding methods increase the detection capacity while a posterior delay compensation methods improve the accuracy of the decomposition. Indicators adapted to the disturbance dynamic are used to characterize its steady-state and transient phases. They are robust to segmentation inaccuracies as well as to steady-state disturbances such as harmonics. Two distinct decision systems are also studied: expert recognition systems and SVM classifiers. During the learning stage, a large simulated event database is used to train both systems. Their performances are evaluated on real events: the type and direction of the measured disturbances are determined with a recognition rate over 98%.

Os Sistemas Supervisórios (SS) executam diversas funções vitais em um processo automatizado e também operam como interface homem-máquina. Os mesmos recebem informações de dispositivos como Controladores Lógicos Programáveis (CLP), inversores de freqüência, etc, e ao mesmo tempo enviam parâmetros de controle fornecidos pelos operadores do processo aos equipamentos de controle. Na operação de SS, a atuação correta e a experiência dos operadores humanos é portanto também vital no controle do processo automatizado. Em recente trabalho(Andrade-2001) de pesquisa na Escola Politécnica da USP foi desenvolvido um Sistema Especialista para operar em tempo real com Sistemas Supervisórios para auxiliar na tomada de decisão dos operadores do sistema. Ao longo do tempo de operação de uma planta automatizada ocorrem novas situações que passam a compor os novos cenários do sistema e também contribuem para o aumento do conhecimento e da experiência dos operadores humanos. Assim sendo, Sistemas Especialistas constantemente devem ser atualizados com novas regras para atender às novas demandas da planta automatizada. Este trabalho de pesquisa apresenta os resultados obtidos com o Sistema Especialista desenvolvido para operar em tempo real com Sistemas Supervisórios, como também o andamento das pesquisas no campo de aprendizado de máquina e mineração de dados com o objetivo de desenvolver e de habilitar Sistemas Especialistas que operam em tempo real com a capacidade de aprender conforme ocorrem eventos durante o funcionamento de uma planta industrial automatizada. / Supervision systems ( SS ) perform diverse vital functions in an automated process and also operate as a man machine interface. The SS receive informations from devices as programmable logical Controllers (PLC), frequency inverters , etc, and at the same time send parameters of control supplied by the process users to the control equipment. In the SS operation, the correct actions and the experience of the human users are therefore vital in the control of the automated process. In a recent research work(Andrade-2001) of the USP Polytechnic School an Expert System was developed to interact in real time with a SS aiming to help in the decision-making process of the system users. During the operation time of an automated plant new situations occur, which come to compose the new system settings and also contribute to the increasing of the human users\' knowledge and experience. Thus, Expert Systems should be constantly brought up to date with these new rules for attend the new demands of the automated plant. This research work shows the results obtained by the expert system, developed to operate in real time with the supervisory systems. It also shows the course of the researches in the fields of machine learning and data mining with the objective of developing and enabling Expert Systems that operate in real time with the capacity to learn events as they happen during the operation of an automated industrial plant.

Os Sistemas Supervisórios (SS) executam diversas funções vitais em um processo automatizado e também operam como interface homem-máquina. Os mesmos recebem informações de dispositivos como Controladores Lógicos Programáveis (CLP), inversores de freqüência, etc, e ao mesmo tempo enviam parâmetros de controle fornecidos pelos operadores do processo aos equipamentos de controle. Na operação de SS, a atuação correta e a experiência dos operadores humanos é portanto também vital no controle do processo automatizado. Em recente trabalho(Andrade-2001) de pesquisa na Escola Politécnica da USP foi desenvolvido um Sistema Especialista para operar em tempo real com Sistemas Supervisórios para auxiliar na tomada de decisão dos operadores do sistema. Ao longo do tempo de operação de uma planta automatizada ocorrem novas situações que passam a compor os novos cenários do sistema e também contribuem para o aumento do conhecimento e da experiência dos operadores humanos. Assim sendo, Sistemas Especialistas constantemente devem ser atualizados com novas regras para atender às novas demandas da planta automatizada. Este trabalho de pesquisa apresenta os resultados obtidos com o Sistema Especialista desenvolvido para operar em tempo real com Sistemas Supervisórios, como também o andamento das pesquisas no campo de aprendizado de máquina e mineração de dados com o objetivo de desenvolver e de habilitar Sistemas Especialistas que operam em tempo real com a capacidade de aprender conforme ocorrem eventos durante o funcionamento de uma planta industrial automatizada. / Supervision systems ( SS ) perform diverse vital functions in an automated process and also operate as a man machine interface. The SS receive informations from devices as programmable logical Controllers (PLC), frequency inverters , etc, and at the same time send parameters of control supplied by the process users to the control equipment. In the SS operation, the correct actions and the experience of the human users are therefore vital in the control of the automated process. In a recent research work(Andrade-2001) of the USP Polytechnic School an Expert System was developed to interact in real time with a SS aiming to help in the decision-making process of the system users. During the operation time of an automated plant new situations occur, which come to compose the new system settings and also contribute to the increasing of the human users\' knowledge and experience. Thus, Expert Systems should be constantly brought up to date with these new rules for attend the new demands of the automated plant. This research work shows the results obtained by the expert system, developed to operate in real time with the supervisory systems. It also shows the course of the researches in the fields of machine learning and data mining with the objective of developing and enabling Expert Systems that operate in real time with the capacity to learn events as they happen during the operation of an automated industrial plant.

Pour automatiser l’harmonisation des bases de données industrielles de navigation sismique, une méthodologie et un logiciel ont été mis en place. La méthodologie d’Automatisation des Mesures de Ressemblance (AMR), permet de modéliser et hiérarchiser les critères de comparaison servant de repères pour l’automatisation. Accompagné d’un ensemble de seuils de tolérance, le modèle hiérarchisé a été utilisé comme filtre à tamis dans le processus de classification automatique permettant de trouver rapidement les données fortement similaires. La similarité est mesurée par un ensemble de métriques élémentaires, aboutissant à des scores numériques, puis elle est mesurée de manière plus globale et contextuelle, notamment suivant plusieurs échelles : entre les attributs, entre les données, et entre les groupes. Ces évaluations de la similarité permettent à la fois au système expert de présenter des analyses précises automatisées et à l’expert géophysicien de réaliser des interprétations multicritères en faisant en environ deux jours le travail qu’il faisait en trois semaines. Les stratégies de classification automatique sont quant à elles adaptables à différentes problématiques, à l’harmonisation des données, mais aussi à la réconciliation des données ou au géo-référencement de documents techniques. Le Logiciel Automatique de Comparaisons (LAC) est une implantation de l’AMR réalisée pour les services de Data Management et de Documentation Technique de TOTAL. L’outil industrialisé est utilisé depuis trois ans, mais n’est plus en maintenance informatique aujourd’hui malgré son usage. Les nouvelles fonctionnalités d'imagerie de base de données qui ont été développées dans cette thèse n'y sont pas encore intégrées, mais devraient permettre une meilleure visualisation des phénomènes. Cette dernière manière de représenter les données, fondée sur la mesure de similarité, permet d’avoir une image assez claire de données lourdes car complexes tout en permettant de lire des informations nécessaires à l’harmonisation et à l’évaluation de la qualité des bases. Ne pourrait-on pas chercher à caractériser, comparer, analyser, gérer les flux entrants et sortants des bases de données, suivre leurs évolutions et tirer des modes d’apprentissage automatique à partir du développement de cette imagerie ? / In order to harmonize industrial seismic navigation data bases, a methodology and a software have been developed. The methodology of Similarity Measurement Automation provides protocols to build a model and a hierarchy for the comparison criteria that shall be used as points of reference for the automation. With its tolerance set of thresholds, the model has been used as a scaled filter within the automatic classification process which aim is to find as quickly as possible very similar data. Similarity is measured by combinations of elementary metrics giving scores, and also by a global and contextual procedure, giving access to three levels of results: similarity between attributes, between individuals, and between groups. Accurate automated analyses of the expert system as well as human interpretations on multiple criteria are now possible thanks to these similarity estimations, reducing to two days instead of three weeks the work of a geophysicist. Classification strategies have been designed to suit the different data management issues, as well as harmonization, reconciliation or geo-referencing. The methodology has been implemented in software for automatic comparisons named LAC, and developed for Data Management and Technical Documentation services in TOTAL. The software has been industrialized and has been used for three years, even if now there is no technical maintenance anymore. The last data base visualization functionalities that have been developed have not been integrated yet to the software, but shall provide a better visualization of the phenomena. This latest way to visualize data is based on similarity measurement and obtains an image of complex and voluminous data clear enough. It also puts into relief information useful for harmonization and data quality evaluation. Would it be possible to characterize, compare, analyze and manage data flows, to monitor their evolution and figure out new machine learning methods by developing further this kind of data base imaging?