O objeto de estudo deste trabalho é a construção de modelos que permitam a estruturação do projeto do controle de sistemas flexíveis de manufatura que considerem não somente estados de operação normal, mas também estados anormais, isto é, em situações de falhas. Entende-se como falha o desvio de pelo menos uma propriedade do sistema que leva o mesmo a um estado de defeito, que por sua vez se define como um comportamento incomum, não projetado, do sistema sob estudo, que finalmente é manifestado como um defeito. Sistemas flexíveis de manufatura são sistemas que executam múltiplos processos visando à produção de diversos bens. O processo é um conjunto de ações de transformação que por sua parte requerem um conjunto de recursos que são compartilhados por outros processos simultaneamente. Sistemas flexíveis de manufatura envolvem um número relativamente grande de componentes, máquinas, equipamentos e operadores humanos, que interagem de maneira diversificada manipulando um grande conjunto de informação e diferentes materiais em ambientes que podem até ser agressivos. Independentemente de qualquer programa de manutenção, falhas são eventos que são possíveis de acontecer em qualquer sistema de tal natureza. Num ambiente ideal, o funcionamento de todos os componentes poderia ser monitorado com o objetivo de detectar as falhas prematuras, mas devido ao custo envolvido, isso se torna inviável. Neste sentido surge o desafio de detectar as falhas a partir da observação do contexto do funcionamento do processo, mediante a monitoração de alguns parâmetros, em geral de fácil acesso, e tomando em consideração manifestações (sintomas) das falhas de um ponto de vista qualitativo. O presente trabalho propõe a utilização de redes Bayesianas para o diagnóstico de falhas em sistemas flexíveis de manufatura. As redes Bayesianas constituem uma ferramenta útil para a representação das relações que existem entre as causas (componentes em estado de falha) e os sintomas (observações anormais do processo). A partir deste modelo, inferências podem ser feitas para o diagnóstico do sistema.Por outro lado, nos últimos anos a rede de Petri tem sido utilizada exitosamente na representação dos aspectos de controle de sistemas produtivos e particularmente de sistemas de manufatura e, desta forma, considera-se aqui tal ferramenta para a modelagem do sistema não só em condições normais de funcionamento como também para a representação do tratamento de falhas, no contexto de um sistema tolerante a anomalias do processo. Especial ênfase é dada à estruturação de uma metodologia que permita a concepção de um procedimento eficaz para a construção de modelos de controle. / The objective of the present work is the construction of models proper for the easy implementation of flexible manufacturing control systems able to handle not only with normal behavioral conditions, but with abnormal (or faulty) behavior as well. A fault is defined as a deviation of at least one system property that drives the system into an error state. An error is defined as an uncommon behavior, not expected from the system functionalities. Flexible manufacturing systems are systems that execute multiple processes for the production of several items in several ways. A process is a sequence of certain transformation tasks that require a set of resources shared simultaneously by multiple processes. In this sense, flexible manufacturing systems are constituted of a relatively great number of devices, machines, equipments and human operators that work together manipulating great quantities of information and materials. This work is usually performed in aggressive environments. So, independent of any maintenance program, faults are events that cannot be totally avoided. In an ideal environment, the monitoring of all components is the way to avoid faults. Nevertheless, due to the cost involved, this is an impossible task. In this context, there is a challenge to properly detect faults from the observation of the systems context, through the monitoring and observation of some parameters in general easy to access, including also qualitative information from operators. In the present work, it is proposed the use of Bayesian networks for the fault diagnosis in flexible manufacturing systems. Bayesian networks constitute a useful tool for the modeling of the causal relation between the causes (faulty components) and the symptoms (manifestations). Based on this model, inference can be done for the system diagnosis task. Additionally, in the last years Petri net has been successfully used for the modeling of control systems of productive systems and particularly, manufacturing control systems. In this work, beyond the use of Petri net for the modeling of normal situations of the system, Petri net is used for the modeling of the fault treatment techniques. This drives the system tolerance to faults. Especial emphasis is laid into methodological issues that allows for the structuration of a systematic procedure proper for the modeling and construction of control systems.

nÃo hà / Esse trabalho propÃe um novo mÃtodo para sÃntese de supervisores em sistemas a eventos discretos, utilizando redes de Petri coloridas, baseado no contexto da teoria de controle supervisÃrio. Como aspecto principal, tem-se a formalizaÃÃo teÃrica e a prova matemÃtica do mÃtodo para sÃntese desses supervisores, denominado RestriÃÃes de Controle sobre Cores Decompostas (RCCD). AlÃmdisso, como consequÃncia, à apresentado ummÃtodo para reduÃÃo do nÃmero de lugares de controle denominado FusÃo de Controladores (FC). Os mÃtodos RCCD e FC sÃo uma extensÃo do trabalho de Moody e Antsaklis, que propÃem, originalmente, um Teorema comprovando a sÃntese de supervisores usando redes de Petri ordinÃrias e, dessa forma, sÃo propostas aqui as extensÃes desse Teorema, para o contexto das redes de Petri coloridas. Nesse caso, explora-se a decomposiÃÃo da matriz de incidÃncia de uma rede de Petri colorida para de…nir lugares de controle que nÃo sÃo con‡itantes em suas aÃÃes, usando o soft CPNTools como ferramenta. A seguir, à realizada a fusÃo desses lugares de controle, obtidos com o RCCD, com a aplicaÃÃo do mÃtodo FC, em que o lugar de controle encontrado, denominado supervisor resultante da fusÃo, atende as mesmas restriÃÃes e contribui com a mesma performance de controle do sistema. / This work proposes a newmethod for synthesis of supervisors in discrete event systems using coloured Petri nets, based in the theory of supervisory control. As a main aspect, there is the theoretical and formal mathematical proof of the method for synthesis of supervisors, called Control Constraints on Decomposed Colors (CCDC). In addition, as a consequence, it is presented a method for reducing the number of control places called Fusion of Controllers (FC). The CCDC and FC methods are an extension of the work of Moody and Antsaklis, they propose, originally a theorem proving the synthesis of supervisors using Petri nets thus are proposed here extensions of this theorem to the context of coloured Petri nets. In this case, it is explored the decomposition of the incidence matrix of a coloured Petri net to de…ne control places that are not con‡icting in their actions, using the soft CPNTools. The following is realized the fusion of these control places obtained with CCDC, with the aplication of the FC method. The obtained place is called the resulting fusion supervisor, where it is found the same constraints and contributes with the same control system performance.

Cette thèse porte sur le diagnostic des défauts dans des Systèmes à Evènements Discrets (SED) pour lesquels l’occurrence des événements est soumise à des contraintes temporelles. Les domaines d’application potentiels sont nombreux et variés allant de la production manufacturière aux systèmes de transport en passant par les réseaux de communication et les systèmes d’information. La complexité croissante de ces systèmes nécessite l’élaboration de méthodes de surveillance de plus en plus efficaces et performantes pour garantir leur sécurité, leur disponibilité ainsi que le maintien de leurs propriétés dans le temps. Les réseaux de Petri Stochastiques Temporisés Partiellement Observés (RdPSTPO) sont utilisés pour modéliser le système ainsi que les défauts qui peuvent l’affecter. L’étude concerne particulièrement les défauts qui se traduisent par la violation des contraintes temporelles. Une exploitation pertinente des informations temporelles, en vue du diagnostic, constitue la contribution majeure de la thèse. En effet, une meilleure utilisation de ces informations permet de discerner avec précision les différents comportements qui expliquent les mesures. De plus, la probabilité d’occurrence des défauts est calculée en fonction des dates des mesures collectées. Deux approches sont développées : la première est dédiée aux défauts ponctuels et la seconde aux dérives lentes dans les systèmes cycliques. / This PhD thesis deals with the fault diagnosis of Discrete Event Systems (DES) for which the occurrence of events is constrained by temporal specifications. The domain of application is large and varied ranging from manufacturing systems to transportation systems, communication networks and information systems. The increasing complexity of these systems requires the development of effective and efficient monitoring methods to ensure their security, availability and the maintenance of their properties over the time. For that purpose, Partially Observed Stochastic Timed Petri Nets (POSTPN) are used to model the system and the different faults that may affect it. The study is particularly concerned with faults that result after the violation of temporal constraints. Our main contribution in this thesis is the relevant exploitation of the timed information for the fault diagnosis. In fact, a better use of these temporal informations allow to discern with precision the different behaviors that are consistent with the measurements. In addition, the probability of faults occurrence is evaluated according to the dates of collected measurements. Two approaches are developed : the first one is used for single faults and the second one for slow drifts in cyclical systems.

<p>Sampled-data supervisory control deals with timed discrete event systems (TDES) where the supervisors are to be implemented as sampled-data controllers. A sampled-data controller views the system as a series of inputs and outputs and is controlled by a periodic clock. It samples its inputs, changes state, and updates its outputs on each clock edge (the tick event). The sampled-data supervisory control framework provides a set of conditions that the TDES system must satisfy to ensure its correct behaviour in order to be implemented as sampled data controllers. A serious limitation for automatic verification of systems is the size of the system's synchronous product. To overcome this limitation, we propose the use of a compositional approach to the verification of sampled-data supervisory control. In this approach, first we recast the required conditions for sampled-data supervisory control in terms of other properties such as language inclusion, nonblocking or controllability, which already have existing compositional methods and algorithms. This makes the sampled-data properties suitable for compositional verification, considerably increasing the size of systems that can be handled using sampled-data supervisory control. We also develop and implement a set of algorithms for the compositional verification of these sampled-data properties. We provide an example of the SD Controlled Flexible Manufacturing System to test our algorithms.</p> / Master of Science (MSc)

Les systèmes avioniques s'étoffent et se complexifient de plus en plus. Avec l'augmentation des capacités de calcul, de nouvelles architectures basées sur le partage de ressources émergent. Effectuer le diagnostic d'un système n'est désormais plus une opération anodine. L'enjeu actuel est donc de mettre en place des techniques de diagnostic performantes tout en optimisant les capacités de monitoring nécessaires.Ce mémoire donne une caractérisation basée modèle d'un système sous diagnostic, puis propose des techniques pour en évaluer les performances de diagnostic, ainsi que celles de son monitoring (relativement à ces performances). Le contexte industriel dans lequel s'inscrit cette thèse amène d'autres contraintes, notamment la prise en compte de la taille des systèmes avioniques à analyser. Cette thèse étudie alors l'applicabilité des techniques introduites dans ce contexte et en propose une adaptation. / Avionics systems become more and more complex. With the improvment of computing possibilities, new architectures based on resources sharing are growing up. Perform diagnosis of a system is no longer a trivial operation. The challenge is to develop efficient techniques of diagnosis while optimizing capabilities of monitoring required.This thesis give a model-based characterization of a system under diagnosis, and proposes techniques to assess diagnostic performances, as well as its monitoring ones (with respect to these diagnostic performances). The industrial context of this thesis brings other constraints, and in particular the need to handle the size of avionics systems to analyze. That thesis then examines the applicability of the introduced techniques to this particular context, and proposes an adaptation.

Le test d’identification d’état d’un système à événement discret (SED) a pour but d’en identifier l’état final, lorsque son état initial est inconnu. Une solution classique à ce problème, en supposant que le SED n’ait pas de sorties observables, consiste à déterminer une séquences de synchronisation, c.à-d., une séquence d’événements d’entrée qui conduit le SED sur un état connu. Ce problème a été résolu dans les années 60’ à l’aide des automates. L’objectif principal de cette thèse est d’utiliser les réseaux de Petri (RdP) pour obtenir une résolution plus optimal de ce problème et pour une plus large classe de systèmes.Initialement, nous montrons que la méthode classique peut être aisément étendue aux RdP synchronisés. Pour cette classe de réseaux non-autonomes, toute transition est associée à un événement d’entrée.L’approche proposée est générale, dans la mesure où elle s’applique à des RdP bornés arbitraires. Cependant, elle engendre le problème d’explosion combinatoire du nombre d’états. Pour obtenir des meilleures solutions, nous considérons une classe spéciale de RdP : les graphes d’état (GdE). Pour ces réseaux, nous considérons d’abord les GdE fortement connexes et proposons des approches pour la construction de SS, qui exploitent les propriétés structurelles du réseau en évitant ainsi une énumération exhaustive de l’espace d’état. Ces résultats s’étendent aux GdE non fortement connexes et à tout RdP synchronisé composé de GdE. Enfin, nous considérons la classe des RdP non bornés et proposons des séquences qui synchronisent le marquage des places non bornées. Une boîte à outils fournit toutes les approches décrites et est appliquée à des différents bancs d’essai. / State-identification experiments are designed to identify the final state of a discrete event system (DES) when its initial state is unknown. A classical solution, assuming the DES has no observable outputs, consists in determining a synchronizing sequence (SS), i.e., a sequence of input events that drives the system to a known state. This problem was essentially solved in the 60’ using automata. The main objective of this thesis is to use Petri nets (PNs) for solving the state-identification problem more efficiently and for a wider class of systems.We start showing that the classical SS construction method based on automata can be easily applied to synchronized PNs, a class of non-autonomous nets where each transition is associated with an input event. The proposed approach is fairly general and it works for arbitrary bounded nets with a complexity that is polynomial with the size of the state space. However, it incurs in the state-space explosion problem.Looking for more efficient solutions, we begin by considering a subclass of PNs called state machines (SMs). We first consider strongly connected SMs and propose a framework for SS construction that exploits structural criteria, not requiring an exhaustive enumeration of the state space of the net. Results are further extended to larger classes of nets, namely non strongly connected SMs and nets containing SM subnets. Finally we consider the class of unbounded nets that describe infinite state systems: even in this case we are able to compute sequences to synchronize the marking of bounded places. A Matlab toolbox implementing all approaches previously described has been designed and applied to a series of benchmarks.

Les systèmes de stockage de l'énergie électrique de forte capacité sont configurés en systèmes matriciels de cellules élémentaires. Les caractéristiques électriques de ces cellules n'évoluent pas toutes de manière identique, diminuant la disponibilité, à court terme par décharge rapide, à long terme en réduisant la durée de vie. Pour améliorer ces performances, des cellules redondantes et des circuits d'équilibrage sont insérés pour assurer une reconfiguration adéquate. Il devrait être possible d'accroître la disponibilité en reconfigurant les connexions internes. Nous comparons deux solutions classiques : série-parallèle (SP) et parallèle-série (PS) avec une nouvelle permettant de redistribuer le courant dans une batterie : le C-3C. Les performances sont évaluées en terme de fiabilité et de disponibilité. Nous proposons également un algorithme de pilotage adapté. La fiabilité est améliorable par redondance. Les cellules supplémentaires seront utilisées pour remplacer des cellules affaiblies. Le système peut également être conçu pour tolérer la défection d'une partie des cellules. Nous démontrons par des diagrammes de fiabilité et des chaînes de Markov que les architectures C-3C et PS présentent le même niveau de fiabilité, supérieur à celui d'une architecture SP. La durabilité des structures peut être améliorée en pilotant la mise en service des ressources disponibles selon différentes stratégies déclinées dans un algorithme de choix fondé sur les États de charge ou les États de Santé. Nous avons modélisé une cellule sous Matlab, en simulant les paramètres de vieillissement et leur évolution dynamique. Ainsi quelle que soit l'architecture, pour peu qu'elle comprenne une part minimale de redondance, une adéquate gestion différentiée des cellules permet une amélioration de la disponibilité de 40%. Par souci de reproductibilité, nous avons également modélisé ces structures par un réseau de Petri coloré, de manière à esquisser l'instrumentation et le dimensionnement de la commande. / High-capacity electrical energy storage system (EESS) are often matrix-organized system with a large number of elementary storage cells. Due to manufactoring tolerances and their individual use, the electrical characteristics of these cells do not evolve in the same way. These imbalances reduce operative dependability, in the short term by contributing to a decrease of the charge-discharge capacity, in the long-term by shortening lifetime. To improve storage performance, redundant cells can be added. It is also possible, in order to increase efficiency of stored energy restitution, to balance electrical characteristics by using energy exchange forced by an adequate configuration. It should therefore be possible to increase long-term operative dependability by reconfiguring internal connections in dynamic mode. Parallel-series (PS) architecture EESS consists of the series association of blocks, made up of several cells connected in parallel. Series-Parallel dual solution (SP) associates strings of cells in parallel. If other architectures are being studied, often requiring several switches per cell to reconfigure the matrix, we propose in this thesis a new architecture, called C3C, satisfying an acceptable level of reliability and distributing current flows. We then compare the classic solutions and the C3C in terms of reliability and the long-term operative dependability and propose a reflection on the possibilities to discrete control aspects to pilot architecture with a suitable control algorithm. The reliability of any structure can be improved by redundancy, with additional cells that will be used either to replace failing cells or temporarily supplemeting the weak ones. The system may also be designed to tolerate the defect of a portion of the cells. We demonstrate by modeling reliability diagrams and Markov chains that the C3C and PS architectures have a much eigher level of reliability than a SP architecture. The sustainability of these structures can also be improved by piloting activating and rest of the available resources according to different strategies in a choice algorithm based on SoC (State of Charge) or SoH (State of Health) of each cell. To do this, we model a cell on Matlab, precisely simulating the aging parameters and their dynamic evolution. It emerges that, whatever the architecture, if it includes a minimal share of redundant cells, an adequate differentiated management of the cells allows an improvement of the long-term operative dependability of nearly 40% on average. In order to study the reconfigurability control of architectures, we propose a model based on Discrete Event Systems through a colored Petri net. Simulation of this model has reinforced the behaviors already identified.

Os sistemas de manufatura há muito vêm sendo objeto de interesse por profissionais e pesquisadores devido à busca de melhores técnicas visando o aumento da produtividade bem como pelo aumento da competitividade empresarial ao longo dos anos. Dentre seus componentes, o sistema de movimentação de materiais merece atenção especial pois, apesar de não aumentar o valor do produto, é responsável por manter o fluxo de materiais entre máquinas, células de manufatura, centros de custos ou mesmo entre empresas, que é imprescindível para qualquer sistema produtivo. Neste contexto, o presente trabalho propõe uma metodologia para a modelagem de sistemas flexíveis de movimentação de materiais e partes em ambiente fabril, focando-se em sistemas cuja movimentação possa ser realizada por VATs (Veículos Autônomos de Transporte). Considerando-se o sistema de movimentação de materiais como sendo um sistema a eventos discretos (SEDs), pode-se empregar técnicas derivadas das Redes de Petri como o PFS (Production Flow Schema) e o E-MFG (Enhanced Mark Flow Graph) na modelagem de tais sistemas. Para tanto, foram introduzidos conceitos de orientação a objetos ao E-MFG de forma a ampliar sua capacidade de modelagem, possibilitando a migração de um paradigma de modelagem orientada a processos para um de modelagem híbrida orientada a processos e a objetos. Como estudo de caso, apresenta-se a modelagem de uma simplificação do sistema de movimentação de materiais da Mercedes Benz do Brasil, situada em São Bernardo do Campo, São Paulo, para a qual se aplica a metodologia proposta. / Manufacturing systems have been object of interest of many professionals and researchers through the years due to the search of better methods for raising goods productivity and managerial competitiveness. Among its components, the material movement system deserves special attention because even not increasing the product value, it is responsible for keeping the flow of materials between machines, manufacturing cells, cost centers and also between companies, which is indispensable for any productive system. In this context, this work proposes a methodology for modeling flexible systems for materials and parts movement in the industrial environment, focusing on systems whose movement can be performed by AGVs (Autonomous Guided Vehicles). Considering the materials movement system as a Discrete Event System (DES), techniques derived from Petri Nets as PFS (Production Flow Schema) and E-MFG (Enhanced Mark Flow Graph) can be used for modeling those systems. To this purpose, concepts of object orientation are introduced to E-MFG in order to increase its modeling capacity, allowing the migration from the process oriented modeling paradigm to an hybrid object and process oriented modeling. A simplification of the materials movement system of the plant of "Mercedes Benz do Brasil", which is located in São Bernado do Campo, São Paulo, was used as a study case to illustrate the methodology presented.

Un Système de production peut être représenté par les systèmes à événements discrets. En dehors de la planification (où les gens travaillent avec des ratios de produits fabriqués par semaine ou par jour), la modélisation pourrait être basée sur les concepts d'événement et d'activités. Un événement correspond à un changement d'état. Une activité est une boîte noire d'encapsulation de ce qui se passe entre deux événements. En utilisant les réseaux de Petri (RdP), les événements sont représentés par les transitions, et les activités par les lieux. Notre travail propose une synthèse de commande par supervision pour les systèmes d'événements discrets modélisés par une classe de réseaux de Petri appelé graphe d'événements. L'objective de cette thèse est de concevoir un superviseur capable d'aider à améliorer la performance de système et de protéger le système en respectant des spécifications données par le fabricant ou le client selon les besoins et les conditions de travail. Pour modéliser ces spécifications, nous avons proposé un nouveau modèle mathématique de contrainte, appelé Contrainte d'Exclusion de Marquage (CEM). La deuxième contribution principale de ma thèse est de synthétiser une technique efficace et simple pour construire un superviseur qui impose le système de respecter des contraintes en évitant l'ensemble des états interdits modélisé par CEM. Nous avons également développé cette synthèse pour résoudre le problème d'existence des événements incontrôlables et des événements inobservables. Parfois, afin d'étudier les aspects liés à la performance, nous devons prendre le temps en considération. Donc, nous avons résolu aussi le problème des événements temporisé en utilisant RdP temporisés soumis à CEM / A manufacturing system may be represented by Discrete Event System (DES). Apart from planning (where people work with ratios of products fabricated per week or per day), any modelling could be based on the concepts of event and activities. An event corresponds to a state change. An activity is a black-box summarizing what is occurring between two events. When using Petri Nets, events are associated with transitions, and activities with places. Our work proposes a supervisory synthesis for Discrete Event System modelled by a class of Petri Net called Marked Graph. The objective of this synthesis is to build a control law that enforces the system to respect a set of given specifications. To model these specifications, we propose new mathematical formulas called Marking Exclusion Constraint (MEC). This model is our first contribution. The Second main contribution of my thesis is to synthesize a computationally efficient technique to build a supervisor that enforces the system to respect the constraints by avoiding a set of forbidden states modelled by MEC specifications. We extend this synthesis technique to solve the problem in the presence of uncontrollable events and unobservable events. Sometimes in order to study the performance aspects, we must take in consideration the time data. Thus we address control synthesis for Timed Discrete Event Systems under MEC specifications by using Timed Petri Nets

Les systèmes automatisés de production (SAP) représentent une classe importante des systèmes industriels qui sont de plus en plus complexes vue le grand nombre d’interaction et d’interconnexion entre leurs différents composants. En conséquence, ils sont plus sensibles aux dysfonctionnements dont les conséquences peuvent être importantes en termes de productivité, de sécurité et de qualité de production. Un défi majeur est alors de développer une approche intelligente qui peut être utilisée pour le diagnostic de ces systèmes afin de garantir leurs suretés de fonctionnement. Dans le cadre de cette thèse, nous nous intéressons seulement au diagnostic des SAP ayant une dynamique discrète. Nous présentons dans le premier chapitre ces systèmes, les dysfonctionnements possibles et la terminologie du diagnostic utilisée. Ensuite, nous présentons un état de l’art de différentes méthodes et approches existantes et aussi une synthèse de ces méthodes. Cette synthèse nous a motivé de choisir une approche à base de donnée qui s’appuie sur une technique d’apprentissage automatique, qui est le raisonnement à partir de cas (RàPC). Pour cela, nous avons présenté dans le deuxième chapitre un état de l’art sur l’apprentissage automatique et ses différentes méthodes en mettant l’accent essentiellement sur le RàPC et ses utilisations pour le diagnostic des systèmes industriels. Cette étude nous a permis de proposer dans le chapitre 3 une approche d’aide au diagnostic qui se base sur le RàPC. Cette approche s’appuie sur une phase hors ligne et une phase en ligne. La phase hors ligne permet de définir un format de représentation de cas et de construire une base de cas normaux (BCN) et une base de cas défaillants (BCD) à partir d’une base de données d’historique. La phase en ligne permet d’aider les opérateurs humains de surveillance à la prise de la décision du diagnostic la plus adéquate. Les résultats des expérimentations sur un système de tri de caisses ont présentés les piliers de cette approche qui résident au niveau du format de représentation de cas proposé et au niveau de la base de cas utilisé. Pour résoudre ces problèmes et améliorer les résultats, un nouveau format de représentation de cas est proposé dans le chapitre 4. Selon ce format et à partir des données issues du système simulé après son émulation en mode normal et fautif, les cas de la base de cas initiale sont construits. Ensuite, une phase de raisonnement et d’apprentissage incrémental est présentée. Cette phase permet non seulement le diagnostic du système surveillé mais aussi d’enrichir la base de cas suite à l’apparition des nouveaux comportements inconnus. Les expérimentations présentées dans le chapitre 5 sur « le plateau tournant » qui est un sous système du système « tri de caisses » ont permis de montrer l’amélioration des résultats et aussi d’évaluer et de comparer les performances de l’approche proposée vis-à-vis certaines approches d’apprentissage automatique et vis-à-vis une approche à base de modèle pour le diagnostic du plateau tournant. / Automated production systems (APS) represents an important class of industrial systems that are increasingly complex given the large number of interactions and interconnections between their different components. As a result, they are more susceptible to malfunctions, whose consequences can be significant in terms of productivity, safety and quality of production. A major challenge is to develop an intelligent approach that can be used to diagnose these systems to ensure their operational safety. In this thesis, we are only interested in the diagnosis of APS with discrete dynamics. We present in the first chapter these systems, the possible malfunctions and the used terminology for the diagnosis. Then, we present a state of the art of the existing methods for the diagnosis of this class of systems and also a synthesis of these methods. This synthesis motivated us to choose a data-based approach that relies on a machine learning technique, which is Case-Based Reasoning (CBR). For this reason, we presented in the second chapter a state of the art on machine learning and its different methods with a focus mainly on the CBR and its uses for the diagnosis of industrial systems. This study allowed us to propose in Chapter 3 a Case Based Decision Support System for the diagnosis of APS. This system is based on an online block and an offline block. The Offline block is used to define a case representation format and to build a Normal Case Base (NCB) and a Faulty Case Base (FCB) from a historical database. The online block helps human operators of monitoring to make the most appropriate diagnosis decision. The experiments results perform on a sorting system presented the pillars of this approach, which reside in the proposed case representation format and in the used case base. To solve these problems and improve the results, a new case representation format is proposed in chapter 4. According to this format and from the data acquired from the simulated system after its emulation in normal and faulty mode, cases of the initial case base are build. Then, a reasoning and incremental learning phase is presented. This phase allows the system diagnosis and the enrichment of the case base following the appearance of new unknown behaviors. The experiments presented in Chapter 5 and perform on the 'turntable' which is a subsystem of the 'sorting system” allowed to show the improvement of the results and also to evaluate and compare the performances of the proposed approach with some automatic learning approaches and with a model-based approach to turntable diagnosis.