La detección de anomalías en procesos industriales es un tema de alto impacto que ha sido analizado y estudiado en diversas áreas de la ingeniería e investigación. La mayor parte de los métodos de detección actualmente disponibles posibilitan el estudio de las irregularidades encontradas en el historial de un proceso, ayudando a extraer información significativa (y a veces crítica) en una amplia variedad de aplicaciones, y convirtiéndose de este modo en parta fundamental e integral de esquemas de reducción de costos tanto humanos como económicos en la industria contemporánea. El objetivo general de este trabajo es desarrollar e implementar un enfoque modular de detección de anomalías, aplicable a procesos industriales multivariados y fundado en el análisis de residuos generados a partir de modelos no paramétricos basados en similitud (similarity-based modeling, SBM). La herramienta consiste principalmente de un sistema de generación automática de modelos SBM, una metodología de estudio de eventos y un algoritmo estadístico de detección. El trabajo realizado se enmarca en un proyecto de colaboración conjunta entre las empresas CONTAC, INGENIEROS LTDA. y ENDESA-CHILE. Gracias a esto, ha sido posible evaluar el sistema propuesto utilizando datos de operación correspondientes a una central termoeléctrica de ciclo combinado perteneciente a la segunda empresa ya mencionada. Las comparaciones en cuanto al desempeño del sistema de modelación implementado permiten concluir que el algoritmo es capaz de generar una representación más apropiada del proceso, basado en el error obtenido con la técnica de modelación SBM, la cual es cercana al 25% del error obtenido con la técnica de modelación lineal en los parámetros. Además, la metodología de estudio de eventos permite detectar correctamente las variables que no aportan al objetivo de detección de un evento en particular, así como también identifica las variables más significativas para lograr tal propósito, reduciendo el número de variables analizadas y con esto, los requerimientos de cómputo de operación en línea. La validación de los resultados entregados por el método de detección de anomalías desarrollado, permite aseverar que la utilización de modelos no-paramétricos tipo SBM, en combinación con la metodología de estudio de eventos y el algoritmo estadístico de detección, resultan eficaces a la hora de generar alarmas y detectar las anomalías estudiadas.

Abstract Automatic three-dimensional scene reconstruction from multiple images is a central problem in geometric computer vision. This thesis considers topics that are related to this problem area. New models and methods are presented for various tasks in such specific domains as camera calibration, image-based modeling and image matching. In particular, the main themes of the thesis are geometric camera calibration and quasi-dense image matching. In addition, a topic related to the estimation of two-view geometric relations is studied, namely, the computation of a planar homography from corresponding conics. Further, as an example of a reconstruction system, a structure-from-motion approach is presented for modeling sewer pipes from video sequences. In geometric camera calibration, the thesis concentrates on central cameras. A generic camera model and a plane-based camera calibration method are presented. The experiments with various real cameras show that the proposed calibration approach is applicable for conventional perspective cameras as well as for many omnidirectional cameras, such as fish-eye lens cameras. In addition, a method is presented for the self-calibration of radially symmetric central cameras from two-view point correspondences. In image matching, the thesis proposes a method for obtaining quasi-dense pixel matches between two wide baseline images. The method extends the match propagation algorithm to the wide baseline setting by using an affine model for the local geometric transformations between the images. Further, two adaptive propagation strategies are presented, where local texture properties are used for adjusting the local transformation estimates during the propagation. These extensions make the quasi-dense approach applicable for both rigid and non-rigid wide baseline matching. In this thesis, quasi-dense matching is additionally applied for piecewise image registration problems which are encountered in specific object recognition and motion segmentation. The proposed object recognition approach is based on grouping the quasi-dense matches between the model and test images into geometrically consistent groups, which are supposed to represent individual objects, whereafter the number and quality of grouped matches are used as recognition criteria. Finally, the proposed approach for dense two-view motion segmentation is built on a layer-based segmentation framework which utilizes grouped quasi-dense matches for initializing the motion layers, and is applicable under wide baseline conditions.

Supply chain management is a complex process requiring the coordination of numerous decisions in the attempt to balance often-conflicting objectives such as quality, cost, and on-time delivery. To meet these and other objectives, a focal company must develop organized systems for establishing and managing its supplier relationships. A reliable, decision-support tool is needed for selecting the best procurement strategy for each supplier, given knowledge of the existing sourcing environment. Supplier segmentation is a well-established and resource-efficient tool used to identify procurement strategies for groups of suppliers with similar characteristics. However, the existing methods of segmentation generally select strategies that optimize performance during normal operating conditions, and do not explicitly consider the effects of the chosen strategy on the supply chain’s ability to respond to disruption. As a supply chain expands in complexity and scale, its exposure to sources of major disruption like natural disasters, labor strikes, and changing government regulations also increases. With increased exposure to disruption, it becomes necessary for supply chains to build in resilience and robustness in the attempt to guard against these types of events. This work argues that the potential impacts of disruption should be considered during the establishment of day-to-day procurement strategy, and not solely in the development of posterior action plans. In this work, a case study of a laser printer supply chain is used as a context for studying the effects of different supplier segmentation methods. The system is examined using agent-based modeling and simulation with the objective of measuring disruption impact, given a set of initial conditions. Through insights gained in examination of the results, this work seeks to derive a set of improved rules for segmentation procedure whereby the best strategy for resilience and robustness for any supplier can be identified given a set of the observable supplier characteristics.

Ce projet utilise l’analyse de réseaux et la modélisation à base d’agents pour examiner des sujets classiquement traités mais qui peuvent maintenant être abordés, grâce aux riches données rencontrées dans le sud-ouest du Colorado et en France méridionale : comment les Gaules et les marchands méditerranéens établissaient leurs partenariats économiques, comment la violence a pu façonner le développement de niveaux divers de leadership, et comment les premiers agriculteurs interagissaient avec leur environnement. Pour écrire cette thèse composée de trois études de cas différents, deux dans le Sud-Ouest des États-Unis et un en France méridionale, nous utilisons des outils élaborés par les sciences de la complexité pour mieux aborder comment les individus de la préhistoire surmontaient les défis liés à l’acquisition de ressources. La modélisation à base d’agents et l’analyse de réseaux (sociaux et trophiques) nous permettront de décrire les processus décisionnels et d’analyser comment le partage de stratégies au sein du groupe peut entraîner une plus grande aptitude des individus à agir au sein du groupe. / This project utilizes network analysis and agent-based modeling to examine long-standing questions that can only now be asked with the rich data provided in southwestern Coloradoand southern France: how Gauls and colonists established economic partnerships, how violence may have shaped the development of multiple levels of leadership, and how earlyf armers interacted with their environments. Writing a dissertation composed of three distinct case studies, two from the U.S. Southwest and one from the south of France, I use tools developed in complexity science to better address how people in the past dealt with challenges related to resource acquisition. Agent-based modeling and network analysis (both social network analysis and trophic network analysis) will allow me to characterize human decision making processes and discuss how sharing of strategies within a group can lead to greater fitness of those in the in-group.

In this work, we have developed user interactive capabilities that allow us to perform segmentation and manipulation of patient-specific geometries required for Computational Fluid Dynamics (CFD) studies, entirely in image domain and within a single platform of ‘IAFEMesh'. Within this toolkit we have added commonly required manipulation capabilities for performing CFD on segmented objects by utilizing libraries like ITK, VTK and KWWidgets. With the advent of these capabilities we can now manipulate a single patient specific image into a set of possible cases we seek to study; which is difficult to do in commercially available software like VMTK, Slicer, MITK etc. due to their limited manipulation capabilities. Levelset representation of the manipulated geometries can be simulated in our flow solver (SCIMITAR-3D) without creating any surface or volumetric mesh. This image-levelset-flow framework offers few advantages. 1) We don't need to deal with the problems associated with mesh quality, edge connectivity related to mesh models, 2) and manipulations like boolean operation result in smooth, physically realizable entities which is challanging in mesh domain. We have validated our image-levelset-flow setup with the known results from previous studies. We have modified the algorithm by Krissian et al. and implemented it for the segmentation of Type-A aortic dissection. Finally, we implemented these capabilities to study the hemodynamics in Type-A aortic dissection. Our image based framework is a first of its kind and the hemodynamic study of Type-A dissection too is first study onto the best of our knowledge.

京都大学 / 0048 / 新制・課程博士 / 博士(経済学) / 甲第22717号 / 経博第620号 / 新制||経||294(附属図書館) / 京都大学大学院経済学研究科経済学専攻 / (主査)教授 関口 倫紀, 教授 若林 直樹, 教授 椙山 泰生 / 学位規則第4条第1項該当 / Doctor of Economics / Kyoto University / DGAM

<div>With the rapid development of clean energy technologies, various bottlenecks on supplies of related critical materials emerged. Since supply chains of critical materials often involved with multiple layers of markets with different characteristics, to better identify bottlenecks and increase critical material availability, it is vital to have better understanding and projection on these markets.</div><div>Agent-based modeling is a bottom-up approach that can imitate heterogenous objects in a changing environment. Therefore, it is an excellent tool to simulate markets with fierce competition and fast revolution. This work demonstrates the application of agent-based modeling by discussing three different topics related to critical material demand and supply induced by clean energy products.</div><div>The first application focused on LED residential lighting market. LED lighting market grew rapidly and introduced potential demand on several critical materials including indium. The work modeled consumers as heterogenous and irrational agents in network purchasing new bulbs based</div>

Content: In most of the tanneries, water and chemicals are added manually in the tanning drum and pH of the float / leather is adjusted. Addition of correct amount of process recipe are necessary for better processing of the hides and minimizing wastage of utility, thereby controlling pollution load in effluent. However, fugitive-emission from process and drains accumulate ammonia, hydrogen-sulphide and volatile organic compounds which contribute bad odor in tannery as well as in wastewater-treatment premises causing problems to occupational health & safety of workers. Both local and supervisory control stations are employed to monitor and accurately manage the unit operations. The objective of this work is to produce consistent quality of leathers and to provide a healthy environment through automatic dosing and odor abatement system. Therefore the entire process control operation is integrated to operate through PLCs with following five modules: i) Water addition module ii) Chemical preparation and dosing system iii) pH monitoring and float recycle system iv) Drum rotation module v) Odor reduction module. In the integrated system, critical and bulk chemicals are stored in bulk storage tanks and are drawn into the load cell (LC) as per process sequence or recipe for feeding into the drums through auxiliary tanks. The float-recycle system helps to remix & heat the float where a pH electrode is housed to monitor pH online. The pH monitoring system adjusts addition of critical chemicals that indicates automatic end point. The contaminated air inside the tannery is sucked and passed through blower and then through bio-filter. The filtration process is based on the principle that VOCs (in the order of 50-200 ppm) and odors can be biologically treated by naturally occurring microbes. The control parameters monitored are: moisture in the bed and uniformity of media (contaminated air or process liquor). The humidity and temperature of inlet media is controlled and contact time with microbes is 10-30 secs. Moisture is controlled to maintain microbial population. A lead in laboratory scale has been developed to measure process variables (PV) considering their spatial distribution in two dimensions. Spatial distribution of process variables inside hides (across cross section) may provide accurate measurement of through reconstruction of image and data driven models using artificial intelligence tools. Computational intelligence is developed for updation of model parameters as that can be used for direct estimation of PV Take-Away: 1. Cleaner production is provided through automation of dosing & pH monitoring using PLC in indegeneous way 2. Pollution Load in exit stream and odor-gas emmision are minimized 3. Artificial Intelligence and data analytics techniques are used in Leather making

No description available.

<p>Innovation is promoted as a means to address global environmental challenges and achieve resilience in the UN Sustainable Development Goals. Innovation allows for adaptation and transformation in socio-ecological systems as part of the adaptive cycle. Within resilience literature, there are myriad definitions of innovation and disagreement about how to motivate diffusion of innovation, making implementation and the sustainability of innovations difficult. Specifically, matching the correct innovation to a given challenge and motivating the adoption of the innovation remains a roadblock to using innovation to address global environmental change. Here we show that there are explicit conflicts among definitions of innovation, and that innovation in the field does not align with some of these definitions. We found that the diverse definitions of innovation show a more complex view of innovation than normative treatment in policy suggests. We also found that several interacting motivations affect long-term participation in certain innovation activities. We discovered that binary views of innovation as either incremental or radical are generally supported in examples of innovation in the field, although some of the most successful examples of innovation better aligned with a continuum view of innovation associated with the adaptive cycle. Our results add to the warm-glow hypothesis that for altruistic tasks, the degree of participation motivated by a warm-glow feeling which can be enhanced by other motivations. Contrary to crowding out theory, our results suggest that monetary incentives result in higher adoption in Malawi where cost of contributing is high. The findings demonstrate the complexity of innovation, the misalignment between policy and practice, and ways in which adoption might be optimized. This research is a starting point to inform discussion about pragmatic innovation typologies. Such a typology could help operationalize the SDGs by framing the innovation dialogue between policy and practice.</p>