No description available.

Fuzzy Cognitive Maps (FCMs) are a widely used, neuro-fuzzy based qualitative approach for the modeling of dynamic systems, which allow for both static and dynamic analyses. They are capable of modeling complex systems with nonlinearities and unknown physical behaviour. FCMs describe a given system by means of concepts connected by quantified cause-effect relationships. This dissertation contributes to the subject of computer-driven generation of FCMs that can be used to perform an accurate dynamic analysis of the modeled system. The dynamic analysis provides insights into the degree of presence, and dependencies between the concepts in successive iterations of the simulation of a given FCM model. Such simulation studies could be used to analyze what-if scenarios in the context of decision support and to perform time series predictions. Two research directions within the framework of FCM development, which concern the learning of FCMs from historical data and an aggregation of FCMs that were proposed by multiple experts, are investigated. Several new automated computational methods for data-driven learning and aggregation of FCMs are introduced and empirically evaluated. These methods utilize real-coded genetic algorithms (RCGA)-based optimization. This choice of the optimization vehicle was motivated by their well-documented efficiency in searching large and continuous search spaces, which are inherent to our problem. Experimental evaluation demonstrates that the proposed RCGA-based learning method outperforms modern existing approaches when the dynamic analysis is considered. A novel divide and conquer-based learning strategy to improve scalability of the RCGA approach, is also proposed. This strategy is shown to be competitive or even better than solutions based on the parallelization of the underlying genetic algorithm. The RCGA-based learning method is further extended to provide improved FCMs when the number of connections of the map is known a priori. Experimental evaluation shows that the density-based learning method outperforms the generic RCGA-based approach when using a relatively accurate density estimate, and that both methods are equivalent when the estimate is inaccurate. In addition, a novel method for the aggregation of multiple input FCMs, is proposed. When compared to existing aggregation approaches, this method provides solutions that are more accurate when dynamic analysis is the objective. / Software Engineering and Intelligent Systems

No description available.

After the Second World War, there was an expression of interest in the study of the future. In order to achieve reliable objectives, several methods of scenario planning were invented. These methods comprise of qualitative and quantitative approaches inherited from their own school of origin. Each approach has advantages and limitations that can be observed by the study of each method in progress. However, mixed approaches often show a weak connection between qualitative and quantitative methods. The use of fuzzy cognitive maps (FCMs), among scenario planning approaches, can overcoming the limitations of qualitative and quantitative techniques. This study focused on the problem of the lack of proper planning of scenarios based on FCMs in the alternative energy sectors, raising the question of whether this situation comes from a shortage of information or limitations of the FCMs. To resolve this problem, this study investigated benefits and limitations of FCM-based scenario planning through the application in the Brazilian photovoltaic sector. The following specific objectives were established: identify the strong and weak points of FCMs; analyze the expansion of these to different areas; identification of the main stakeholders used to develop FCM-based scenarios; application of the method in the Brazilian solar energy sector. We identified an increasing number of studies on FCMs in several new areas not considered in previous studies. This suggests that experts are expanding frontiers and recognizing the interdisciplinary potential of FCMs and their robustness of solving diverse kinds of problems. The main contribution of the research is to present the benefits and limitations of FCM-based scenario planning. For that, a scenario planning method was presented and the empirical evidence of its effectiveness was presented in the context of the Brazilian photovoltaic solar energy sector. / Após a Segunda Guerra Mundial, houve um crescente de interesse no estudo do futuro. Para alcançar objetivos confiáveis, foram inventados vários métodos de planejamento de cenários. Esses métodos possuem abordagens qualitativas e quantitativas herdadas de sua própria escola de origem. Cada abordagem possui vantagens e limitações que podem ser observadas pelo estudo de cada método. No entanto, abordagens mistas muitas vezes mostram uma conexão fraca entre métodos qualitativos e quantitativos. O uso de fuzzycognitivemaps (FCMs), no planejamento de cenários, pode superar as limitações das técnicas qualitativas e quantitativas. Este estudo centrou-se no problema da falta de planejamento adequado de cenários baseados em FCMs nos setores de energia alternativa, levantando a questão de saber se esta situação vem da escassez de informações ou das limitações dos FCMs no planejamento estratégico. Para resolver este problema, este estudo investigou os benefícios e as limitações do planejamento de cenários baseado em FCM através da aplicação no setor fotovoltaico brasileiro. Foram estabelecidos os seguintes objetivos específicos: identificar os pontos fortes e fracos dos FCMs; analisar a expansão destes para diferentes áreas; identificar os principais stakeholders utilizados para desenvolver cenários baseados em FCMs; aplicar o método no setor de energia solar brasileira. Identificamos um número crescente de estudos sobre FCMs em várias novas áreas não consideradas em estudos anteriores. Isso sugere que os especialistas estão expandindo as fronteiras e reconhecendo o potencial interdisciplinar dos FCMs e sua robustez na resolução de diversos tipos de problemas. A principal contribuição desta pesquisa é apresentar os benefícios e as limitações do planejamento de cenário baseado em FCM. Para isso, foi apresentado um método de planejamento de cenários e a evidência empírica de sua eficácia foi apresentada no contexto do setor de energia solar fotovoltaica brasileira.

After the Second World War, there was an expression of interest in the study of the future. In order to achieve reliable objectives, several methods of scenario planning were invented. These methods comprise of qualitative and quantitative approaches inherited from their own school of origin. Each approach has advantages and limitations that can be observed by the study of each method in progress. However, mixed approaches often show a weak connection between qualitative and quantitative methods. The use of fuzzy cognitive maps (FCMs), among scenario planning approaches, can overcoming the limitations of qualitative and quantitative techniques. This study focused on the problem of the lack of proper planning of scenarios based on FCMs in the alternative energy sectors, raising the question of whether this situation comes from a shortage of information or limitations of the FCMs. To resolve this problem, this study investigated benefits and limitations of FCM-based scenario planning through the application in the Brazilian photovoltaic sector. The following specific objectives were established: identify the strong and weak points of FCMs; analyze the expansion of these to different areas; identification of the main stakeholders used to develop FCM-based scenarios; application of the method in the Brazilian solar energy sector. We identified an increasing number of studies on FCMs in several new areas not considered in previous studies. This suggests that experts are expanding frontiers and recognizing the interdisciplinary potential of FCMs and their robustness of solving diverse kinds of problems. The main contribution of the research is to present the benefits and limitations of FCM-based scenario planning. For that, a scenario planning method was presented and the empirical evidence of its effectiveness was presented in the context of the Brazilian photovoltaic solar energy sector. / Após a Segunda Guerra Mundial, houve um crescente de interesse no estudo do futuro. Para alcançar objetivos confiáveis, foram inventados vários métodos de planejamento de cenários. Esses métodos possuem abordagens qualitativas e quantitativas herdadas de sua própria escola de origem. Cada abordagem possui vantagens e limitações que podem ser observadas pelo estudo de cada método. No entanto, abordagens mistas muitas vezes mostram uma conexão fraca entre métodos qualitativos e quantitativos. O uso de fuzzycognitivemaps (FCMs), no planejamento de cenários, pode superar as limitações das técnicas qualitativas e quantitativas. Este estudo centrou-se no problema da falta de planejamento adequado de cenários baseados em FCMs nos setores de energia alternativa, levantando a questão de saber se esta situação vem da escassez de informações ou das limitações dos FCMs no planejamento estratégico. Para resolver este problema, este estudo investigou os benefícios e as limitações do planejamento de cenários baseado em FCM através da aplicação no setor fotovoltaico brasileiro. Foram estabelecidos os seguintes objetivos específicos: identificar os pontos fortes e fracos dos FCMs; analisar a expansão destes para diferentes áreas; identificar os principais stakeholders utilizados para desenvolver cenários baseados em FCMs; aplicar o método no setor de energia solar brasileira. Identificamos um número crescente de estudos sobre FCMs em várias novas áreas não consideradas em estudos anteriores. Isso sugere que os especialistas estão expandindo as fronteiras e reconhecendo o potencial interdisciplinar dos FCMs e sua robustez na resolução de diversos tipos de problemas. A principal contribuição desta pesquisa é apresentar os benefícios e as limitações do planejamento de cenário baseado em FCM. Para isso, foi apresentado um método de planejamento de cenários e a evidência empírica de sua eficácia foi apresentada no contexto do setor de energia solar fotovoltaica brasileira.

L'énergie est fondamentale pour maintenir le confort et façonne notre vie moderne. Avec la demande excédentaire en énergie, les systèmes de gestion de l’énergie résidentielle apparaissent avec le temps. Ils visent à réduire ou moduler la consommation d’énergie tout en maintenant un niveau de confort acceptable. Des systèmes efficaces de gestion de l'énergie domestique devraient intégrer une représentation comportementale d'un système domestique, y compris les habitants. Il établit des relations entre différentes variables environnementales et des phénomènes hétérogènes présents dans une maison. Par conséquent, ces systèmes sont complexes à construire et à comprendre pour les habitants. Pour cette raison, les concepteurs ont essayé d'automatiser autant que possible les systèmes de CVC, les éclairages ... afin de promouvoir le concept de "faire à la place". Cela était justifié car il était presque impossible d'impliquer les occupants et de créer une relation entre les occupants et les systèmes énergétiques. Ce concept crée différents problèmes car les occupants sont détachés du système énergétique et ne comprennent pas ses fonctionnalités ni son fonctionnement.Pour surmonter cette difficulté, ce travail met en avant le concept de "faire avec" en essayant d'impliquer l'occupant dans la boucle avec son système de gestion de l'énergie. C'est là que l'explication est nécessaire pour permettre aux occupants de découvrir les connaissances du système énergétique et de développer leur capacité à comprendre comment le système fonctionne et pourquoi il recommande différentes actions. L'explication est le moyen de découvrir de nouvelles connaissances et, par conséquent, d'impliquer les occupants. Pour les humains, l'explication joue un rôle important dans la vie. C'est l'un des principaux outils d'apprentissage et de compréhension. Il est même utilisé dans la communication et les aspects sociaux. Les gens ont tendance à l'utiliser en plus d'apprendre à montrer leurs connaissances sur un sujet pour gagner la confiance des autres ou pour clarifier une situation. Mais générer des explications n’est pas une tâche facile. C'est l'un des problèmes scientifiques récurrents de plusieurs décennies. Les explications ont de nombreuses formes, types et niveaux de clarté. Cette étude se concentre sur les explications causales. Comme il s’agit de la forme d’explication la plus intuitive à comprendre par les occupants, elle est conçue pour transférer les connaissances issues de systèmes complexes tels que les modèles énergétiques. Le défi scientifique est de savoir comment construire des explications de causalité pour les habitants à partir d’un flux de données de capteurs observées. / Energy is fundamental to maintain comfort and it shapes our modern life. With the excess demand for energy, home energy management systems are appearing with time. They aim at reducing or modulating energy consumption while keeping an acceptable level of comfort. Efficient home energy management systems should embed a behavioral representation of a home system, including inhabitants. It establishes relationships between different environmental variables and heterogeneous phenomena present in a home. Therefore, those systems are complex to build and to understand for inhabitants. For this reason, the designers did try to automatize as much as possible the HVAC systems, the lightings ... so they promoted the concept of “doing instead”. This was justified as it was nearly impossible to implicate occupants and to create a relation between occupants and energy systems. This concept does create different problems as occupants are detached from the energy system and they don’t understand its functionality nor how it is working.To overcome this difficulty this work promotes the concept of “doing with” as it tries to implicate the occupant in the loop with their energy management system. This is where the explanation is needed to allow occupants to discover the knowledge in the energy system and to develop their capacity of understanding how the system is working and why it is recommending different actions. The explanation is the way to discover new knowledge and consequently, to involve occupants. For humans, explanation plays an important role in life. It is one of the main tools for learning and understanding. It is even used in communication and social aspects. People tend to use it besides learning to show their knowledge about a subject to gain the confidence of others or to clarify a situation. But generating explanations is not an easy task. It is one of the ongoing scientific problems from several decades. Explanations have numerous forms, types, and level of clearness. This study is focusing on the causal explanations. As it is the most intuitive form of explanation to be understood by occupants and is adapted to transfer the knowledge from complex systems like energy models. The scientific challenge is how to construct causal explanations for the inhabitants from a flow of observed sensor data.

Στην παρούσα διπλωματική εργασία παρουσιάζονται τα ιατρικά συστήματα λήψης απόφασης (MDSS) και αρχιτεκτονικές ανάπτυξή τους. Πραγματεύεται τις έννοιες του ευφυούς ελέγχου και της ασάφειας για να καταλήξει στον όρο Ασαφή Γνωστικά Δίκτυα(FCΜ). Αφού περιγράφεται αναλυτικά η ανάπτυξή, ο καθορισμός των παραμέτρων και οι μεθοδολογίες εκμάθησης ενός Ασαφούς Γνωστικού Δικτύου, καταλήγει τελικά στην εφαρμογή τους στον χώρο της ιατρικής. Τέλος, ακολουθεί το μοντέλο ενός Ασαφούς Ελεγκτή για ιατρικές εφαρμογές και η ανάπτυξη ενός MDSS για την εύρεση Κάκωσης Γόνατος με αρχιτεκτονικές Ανταγωνιστικού Ασαφούς Γνωστικού Δικτύου (CFCΜ). / The thesis represents the medical decision support systems (MDSS) and their architecture. Starting with the concepts of intelligent control and Fuzzy Cognitive Maps (FCM), it describes in detail the development, the setting parameters, and the learning methods of FCMs, with the purpose of their application into the field of medicine. Finally, it illustrates the model of a Fuzzy Controller for medical applications and the development of an MDSS for finding knee injury with the architecture of Competitive FCMs (CFCM).

Abstract Bleaching is an essential process in chemical pulp production for better pulp brightness and longer life expectancy. However, it causes costs such as chemicals, energy, equipment, and loss of yield. Non-linear reactions and several process variables, with interactions, make large plants complicated to model and optimize. As an expensive process bleaching has been a natural target of optimization, but there is still the need to either improve these methods or consider the optimization problem from a new point of view. The aim of this thesis was to develop production optimization methods for pulp bleaching, so that they are practical, usable on-line, easy to tune, and transferable. According to our assumption, neural networks could provide a practical optimization method by combining analytical knowledge with real data. In this kind of problem, the load sharing concept, recognizing interactions in chemical usage and the serial multi-stage nature of the process can simplify the task. The related work in bleaching optimization was studied as well as multi-stage serial process solving in principle, related optimization methods and especially neural networks in optimization. The data were collected during normal mill operation and modeled using neural networks. Optimization was performed based on visualizing the neural network models. The results showed that backpropagation neural networks are capable of modeling parts of the bleach plant and also the entire bleaching operation to such an extent that they are useful in the optimization. The modeling and the tuning can be performed without a profound knowledge of the system, but the process is slower and less reliable. Moving a trained neural network to another mill is inadvisable. It is more reasonable just to transfer the knowledge of variables and network structure. The important factor in on-line production optimization is the stabilization of the disturbances and a well-controlled operation towards a more economical state. Generally, more than half of the total chemicals should be used in the first bleaching stage D0 and the remaining load should be divided so that the dosage at the D1 is about 30% higher than in the D2 stage.

Food systems are complex social-ecological systems. Currently, they are the source of large-scale health problems and environmental impacts, and there is widespread agreement that transformative change is needed. Scenarios are useful tools for directing such change, as they provide engaging future visions that work well with complex systems.  This thesis is a part of Mistra Food Futures, a platform for a sustainable transformation of the Swedish food system, where scenarios for Swedish food futures are being developed. The thesis purpose is to contribute to the scenario development by the use of systems mapping and semi-quantitative scenario modelling. The thesis builds on four scenario narratives previously developed by Mistra Food Futures researchers. During the thesis process, these scenario narratives were re-interpreted as Causal Loop Diagrams. The diagrams were then used as the basis for constructing a food system model in the form of a Fuzzy Cognitive Map. Simulations were run to investigate the conditions under which the scenarios could be reproduced by modelling. The modelling uncovered several system dynamics: the competition or shared interests of different types of agriculture; the system impacts of novel foods; the vulnerabilities of localised food systems; the importance of food culture; and the interactions of environmental policy with farming systems. Another finding was system attractors where scenarios mix, and these are presented as alternative scenarios. The thesis contributes to the scenario development by making relationships, system feedbacks, and drivers explicit by systems mapping, as well as providing a user-friendly model that can be used for further system exploration. The analysis of specific dynamics can be used to inform upcoming scenario iterations, and alternative scenarios can be used to maintain analytical depth when scenario interactions are discussed. The process also provides a demonstration of the use of Fuzzy Cognitive Maps in scenario modelling.

No description available.