Summary Objective - The objective of this research study is to develop improvement proposals for an ordering handling system of uncontrolled packaging materials. By analysing data from related products in combination with observations and interviews the following research questions were answered to reach the purpose of the study: How can reordering points and inventory handling of uncontrolled packaging materials be structured? What would be the advantage of implement and plan uncontrolled packaging materials in an ERP system? Method - By studying a business phenomenon without firstly determine a theoretical background, the approach of this report has been through inductive research. By first observing the problem and connecting it to related theories the problem could be analysed and processed according to a theoretical approach. Results - The findings of the study show that data of uncontrolled materials can be derived through the connection of relevant products by using a breakdown structure. By this approach data and forecasts of related products could be summarized and used as a calculation basis for a reordering point system. The advantages of implementing reordering points of uncontrolled materials into an ERP system resulted in a reduction of non-value adding activities, reduction of maximal inventory, reduced tied-up capital, and to a controlled ordering handling. Implications - To further the study area of uncontrolled materials the researchers suggest that studies within other areas than packaging material is examined with a similar approach. Parts of the study, such as the combined demand and productmapping, could be applied in projects with more complex structures to explore the limitations of the method. This would increase the knowledge base on how to structure and control materials that do not follow any system. Furthermore, the scope of the report has been to focus on high volume packaging materials and therefore it would be interesting to see the approach applied to low volume products or materials. Restrictions - As data of this study is created by connection of related products it entails that similar circumstances exist if the project is to be replicated. Furthermore, the study is solely focused on the inventory department which could be a delimitation. Charts and data illustrations – In the report charts and data are manipulated according to the confidentiality agreement with the case company. The provided examples illustrate the relation between different factors although does not reveal any true data derived from the case company.

A crescente perda mundial de biodiversidade tem sido uma das preocupações da comunidade científica internacional que motivaram a criação em 1992 da Convenção sobre a Diversidade Biológica, tornando-se um tratado de compromisso aceito pelos governantes de 187 países e pela Comunidade Européia. A redução da biodiversidade, devido a vários fatores, como a ação antrópica e o aquecimento global, compromete a capacidade do planeta de sustentação da vida humana em face do esgotamento dos recursos e serviços por ela prestados. A conservação e uso sustentável da biodiversidade passa necessariamente pela aquilatação e conhecimento das espécies. Entre essas espécies as abelhas polinizadoras têm merecido especial atenção, pois a polinização das plantas é um serviço de ecossistema muito importante. Cerca de três quartos das mais de 240 mil espécies de plantas do mundo dependem de polinizadores e estima-se que as abelhas sejam responsáveis por mais de 70% do serviço global de polinização. Existem quase 20.000 espécies de abelhas descritas no mundo. No Brasil são conhecidas quase 400 espécies de abelhas (cerca de 300 são abelhas sem ferrão) e o número estimado é de mais de 3.000 espécies. No entanto, este enorme esforço taxonômico a ser realizado pode estar comprometido pelo impedimento taxonômico reconhecido na Declaração de Darwin de 1988 pela ONU. Uma contribuição para minimizar o impedimento taxonômico pode ser dada pelo desenvolvimento de sistemas automatizados de apoio à decisão de identificação. Os sistemas de identificação existentes são proprietários, ou foram desenvolvidos para grupo específicos de espécies e muitas vezes não são integrados. Este trabalho com foco na identificação de abelhas, notadamente nas abelhas sem ferrão, que apresentam redução da venação das asas, apresenta um modelo de sistema baseado em computador para automatizar o processo de identificação de abelhas com uma abordagem de reconhecimento de padrões. Um modelo de sistema denominado ABeeS (Automated Bee Identification System) incorpora o conhecimento especializado para o reconhecimento automatizado de abelhas usando a imagem das asas. O modeloproposto apresenta as funcionalidades de um sistema de identificação de abelhas com o modelo de Caso de Uso e o fluxo de dados entre as atividades do processo de identificação com o modelo do Fluxo de Dados. Um modelo de banco de dados denominado Banco Entomológico de Espécies de Abelhas (BEE) foi proposto para armazenamento de resultados, treinamento e otimização do sistema ABeeS. Para levantar requisitos e avaliar a proposta foram desenvolvidos protótipos de partes do modelo no Labview, um ambiente de programação gráfica, que disponibiliza uma plataforma de visão computacional para aplicações de reconhecimento de padrões usando o método de correspondência de padrões (pattern matching). Um protótipo envolveu uma ferramenta para a definição dos marcos anatômicos nas imagens-padrão. Outro envolveu os resultados da extração automática dos marcos anatômicos e evidenciou que mostram que a capacidade do ABeeS em localizar automaticamente os gabaritos (template) dos marcos anatômicos na imagem da asa em análise depende do conhecimento especializado transferido para o sistema. Este conhecimento contribui para a seleção de uma região do entorno do marco anatômico para formação da imagem-gabarito. A definição precisa da área é muito importante para a acurácia do reconhecimento automatizado do marco anatômico. O ajuste dos parâmetros de treinamento e a qualidade da imagem da asa de abelha são determinantes para extração das características corretas. Parte integrante do modelo são algoritmos de classificação supervisionados, como o FNN4Bees desenvolvido no Laboratório de Automação Agrícola da POLI-USP, e que apresentou resultados satisfatórios. Este trabalho contribui com uma sistematização do processo de identificação de abelhas, servindo de guia para usuários dessa técnica, e o modelo obtido poderá ser utilizado para a implementação de um sistema real, na continuidade dos trabalhos. / The growing worldwide loss of biodiversity has been a concern to the international scientific community that motivated the creation of the Convention on Biological Diversity in 1992, which turned into a commitment treaty accepted by governments of 187 countries and the European Community. The reduction of biodiversity due to several factors, such as anthropic action and global warming, compromises the ability of the planet to sustain human life in face of the exhaustion of the resources and services it provides. Conservation and the sustainable use of biodiversity necessarily involve species knowledge and assessment. Among these species, pollinating bees have deserved special attention because plant pollination is an important ecosystem service. Nearly three quarters of more than 240,000 plant species depend on pollinators around the world and bees are estimated to account for more than 70% of the overall pollination service. There are nearly 20,000 bee species described in the world. Brazil is known to have nearly 400 bee species (about 300 are stingless bees) and the estimated number is over 3,000 species. However, this huge taxonomic effort to be performed may be compromised by the taxonomic impediment recognized by the UN in the Darwin Declaration in 1988. A contribution to minimize the taxonomic impediment may be given by the development of automated systems to support classification decision. Existing identification systems are proprietary, or were developed for specific groups of species and are often non-integrated. This study focused on bee classification, especially stingless bees that have reduced wing venation; a model for a computer-based system to automate the bee identification process using a pattern recognition approach is presented. A model of the system called ABeeS (Automated Bee Identification System) incorporates the expertise for the automated recognition of bees using wing images. The model presents the functionalities of the bee identification system with the Use Case model and data flow between the activities of the identification process with the Data Flow model. A database model called Bee Entomological Database (BEE) has been proposed for storage of results, training and for ABeeS system optimization. For requirements elicitation and evaluation, the proposal prototypes of parts of the model were developed in Labview, a graphical programming environment that provides a platform for computer vision applications of pattern recognition using the pattern matching method. One prototype uses a tool for defining the landmarks in the template-images. Another uses the results of the automatic extraction of landmarks that show the ability of ABeeS to automatically locate the templates of landmarks in the wing image under analysis, which depends on the expertise transferred to the system. This knowledge contributes to the selection of a region surrounding the anatomical landmark for the formation of the template. The precise definition of the area is very important for the accuracy of automated recognition of anatomical landmark. The adjustment of training parameters and image quality of the bee wing are crucial for extracting the right features. Part of the model is supervised classification algorithms, such as FNN4Bees developed in the Agricultural Automation Laboratory of POLI-USP, and presents satisfactory results. This work contributes to the systematization of the bee identification process, serving as a guide for users of this technique, and the model obtained can be used to implement a real system, in further work.

A crescente perda mundial de biodiversidade tem sido uma das preocupações da comunidade científica internacional que motivaram a criação em 1992 da Convenção sobre a Diversidade Biológica, tornando-se um tratado de compromisso aceito pelos governantes de 187 países e pela Comunidade Européia. A redução da biodiversidade, devido a vários fatores, como a ação antrópica e o aquecimento global, compromete a capacidade do planeta de sustentação da vida humana em face do esgotamento dos recursos e serviços por ela prestados. A conservação e uso sustentável da biodiversidade passa necessariamente pela aquilatação e conhecimento das espécies. Entre essas espécies as abelhas polinizadoras têm merecido especial atenção, pois a polinização das plantas é um serviço de ecossistema muito importante. Cerca de três quartos das mais de 240 mil espécies de plantas do mundo dependem de polinizadores e estima-se que as abelhas sejam responsáveis por mais de 70% do serviço global de polinização. Existem quase 20.000 espécies de abelhas descritas no mundo. No Brasil são conhecidas quase 400 espécies de abelhas (cerca de 300 são abelhas sem ferrão) e o número estimado é de mais de 3.000 espécies. No entanto, este enorme esforço taxonômico a ser realizado pode estar comprometido pelo impedimento taxonômico reconhecido na Declaração de Darwin de 1988 pela ONU. Uma contribuição para minimizar o impedimento taxonômico pode ser dada pelo desenvolvimento de sistemas automatizados de apoio à decisão de identificação. Os sistemas de identificação existentes são proprietários, ou foram desenvolvidos para grupo específicos de espécies e muitas vezes não são integrados. Este trabalho com foco na identificação de abelhas, notadamente nas abelhas sem ferrão, que apresentam redução da venação das asas, apresenta um modelo de sistema baseado em computador para automatizar o processo de identificação de abelhas com uma abordagem de reconhecimento de padrões. Um modelo de sistema denominado ABeeS (Automated Bee Identification System) incorpora o conhecimento especializado para o reconhecimento automatizado de abelhas usando a imagem das asas. O modeloproposto apresenta as funcionalidades de um sistema de identificação de abelhas com o modelo de Caso de Uso e o fluxo de dados entre as atividades do processo de identificação com o modelo do Fluxo de Dados. Um modelo de banco de dados denominado Banco Entomológico de Espécies de Abelhas (BEE) foi proposto para armazenamento de resultados, treinamento e otimização do sistema ABeeS. Para levantar requisitos e avaliar a proposta foram desenvolvidos protótipos de partes do modelo no Labview, um ambiente de programação gráfica, que disponibiliza uma plataforma de visão computacional para aplicações de reconhecimento de padrões usando o método de correspondência de padrões (pattern matching). Um protótipo envolveu uma ferramenta para a definição dos marcos anatômicos nas imagens-padrão. Outro envolveu os resultados da extração automática dos marcos anatômicos e evidenciou que mostram que a capacidade do ABeeS em localizar automaticamente os gabaritos (template) dos marcos anatômicos na imagem da asa em análise depende do conhecimento especializado transferido para o sistema. Este conhecimento contribui para a seleção de uma região do entorno do marco anatômico para formação da imagem-gabarito. A definição precisa da área é muito importante para a acurácia do reconhecimento automatizado do marco anatômico. O ajuste dos parâmetros de treinamento e a qualidade da imagem da asa de abelha são determinantes para extração das características corretas. Parte integrante do modelo são algoritmos de classificação supervisionados, como o FNN4Bees desenvolvido no Laboratório de Automação Agrícola da POLI-USP, e que apresentou resultados satisfatórios. Este trabalho contribui com uma sistematização do processo de identificação de abelhas, servindo de guia para usuários dessa técnica, e o modelo obtido poderá ser utilizado para a implementação de um sistema real, na continuidade dos trabalhos. / The growing worldwide loss of biodiversity has been a concern to the international scientific community that motivated the creation of the Convention on Biological Diversity in 1992, which turned into a commitment treaty accepted by governments of 187 countries and the European Community. The reduction of biodiversity due to several factors, such as anthropic action and global warming, compromises the ability of the planet to sustain human life in face of the exhaustion of the resources and services it provides. Conservation and the sustainable use of biodiversity necessarily involve species knowledge and assessment. Among these species, pollinating bees have deserved special attention because plant pollination is an important ecosystem service. Nearly three quarters of more than 240,000 plant species depend on pollinators around the world and bees are estimated to account for more than 70% of the overall pollination service. There are nearly 20,000 bee species described in the world. Brazil is known to have nearly 400 bee species (about 300 are stingless bees) and the estimated number is over 3,000 species. However, this huge taxonomic effort to be performed may be compromised by the taxonomic impediment recognized by the UN in the Darwin Declaration in 1988. A contribution to minimize the taxonomic impediment may be given by the development of automated systems to support classification decision. Existing identification systems are proprietary, or were developed for specific groups of species and are often non-integrated. This study focused on bee classification, especially stingless bees that have reduced wing venation; a model for a computer-based system to automate the bee identification process using a pattern recognition approach is presented. A model of the system called ABeeS (Automated Bee Identification System) incorporates the expertise for the automated recognition of bees using wing images. The model presents the functionalities of the bee identification system with the Use Case model and data flow between the activities of the identification process with the Data Flow model. A database model called Bee Entomological Database (BEE) has been proposed for storage of results, training and for ABeeS system optimization. For requirements elicitation and evaluation, the proposal prototypes of parts of the model were developed in Labview, a graphical programming environment that provides a platform for computer vision applications of pattern recognition using the pattern matching method. One prototype uses a tool for defining the landmarks in the template-images. Another uses the results of the automatic extraction of landmarks that show the ability of ABeeS to automatically locate the templates of landmarks in the wing image under analysis, which depends on the expertise transferred to the system. This knowledge contributes to the selection of a region surrounding the anatomical landmark for the formation of the template. The precise definition of the area is very important for the accuracy of automated recognition of anatomical landmark. The adjustment of training parameters and image quality of the bee wing are crucial for extracting the right features. Part of the model is supervised classification algorithms, such as FNN4Bees developed in the Agricultural Automation Laboratory of POLI-USP, and presents satisfactory results. This work contributes to the systematization of the bee identification process, serving as a guide for users of this technique, and the model obtained can be used to implement a real system, in further work.

In this dissertation, we proposed solutions on novel sensor deployment and diagnoser design to efficiently detect stochastic faults in PLC based automated systems First, a fuzzy quantitative graph based sensor deployment was called upon to model cause-effect relationship between faults and sensors. Analytic hierarchy process (AHP) was used to aggregate the heterogeneous properties between sensors and faults into single edge values in fuzzy graph, thus quantitatively determining the fault detectability. An appropriate multiple objective model was set up to minimize fault unobservability and cost while achieving required detectability performance. Lexicographical mixed integer linear programming and greedy search were respectively used to optimize the model, thus assigning the sensors to faults. Second, a diagnoser based on real time fuzzy Petri net (RTFPN) was proposed to detect faults in discrete manufacturing systems. It used the real time PN to model the manufacturing plant while using fuzzy PN to isolate the faults. It has the capability of handling uncertainties and including industry knowledge to diagnose faults. The proposed approach was implemented using Visual Basic, and tested as well as validated on a dual robot arm. Finally, the proposed sensor deployment approach and diagnoser were comprehensively evaluated based on design of experiment techniques. Two-stage statistical analysis including analysis of variance (ANOVA) and least significance difference (LSD) were conducted to evaluate the diagnosis performance including positive detection rate, false alarm, accuracy and detect delay. It illustrated the proposed approaches have better performance on those evaluation metrics. The major contributions of this research include the following aspects: (1) a novel fuzzy quantitative graph based sensor deployment approach handling sensor heterogeneity, and optimizing multiple objectives based on lexicographical integer linear programming and greedy algorithm, respectively. A case study on a five tank system showed that system detectability was improved from the approach of signed directed graph's 0.62 to the proposed approach's 0.70. The other case study on a dual robot arm also show improvement on system's detectability improved from the approach of signed directed graph's 0.61 to the proposed approach's 0.65. (2) A novel real time fuzzy Petri net diagnoser was used to remedy nonsynchronization and integrate useful but incomplete knowledge for diagnosis purpose. The third case study on a dual robot arm shows that the diagnoser can achieve a high detection accuracy of 93% and maximum detection delay of eight steps. (3) The comprehensive evaluation approach can be referenced by other diagnosis systems' design, optimization and evaluation.

During the spring of 2014, Johan Kooiker and Kashira Roil Beto, students in mechanical engineering at Linnaeus University carried out a degree project together with Husqvarna construction products Sweden AB. The topic of this project was "product development" to develop an automated system for assembling cold pressed segments. The project began with studying the current situation to get a clearer picture of the problem and finding out the requirements. The main requirements were studied and implanted in behaviour descriptions of the system. Later on ideas and concepts were generated, which then through rating and screening led to the final concept. The final concept was divided into subsystems and developed in details. The detail work included 3D cad models of unique components and selection of standard component. The final step was to build simple prototypes of the subsystems to test their functions. The prototypes of the subsystems accomplished the main requirements of the company and that led to future investment in final system.

Catfish aquaculture is the largest cultured food fish industry in the United States, accounting for approximately $375 million in sales annually, with Mississippi leading the industry with 36,200 surface acres of production. However, infectious diseases such as enteric septicemia of catfish (ESC) are decreasing production efficiencies, creating losses of $40-60 million annually. Live-attenuated oral ESC vaccines are effective in preventing ESC infections, but have not been widely adopted by the catfish industry due to the lack of a system to administer the oral vaccine at the scale seen in commercial catfish production systems. An automated system was developed to administer a dosage of 220.5 ml of a live-attenuated ESC vaccine per kg of catfish feed, adapting commercial catfish feeder design to include a screw conveyor for mixing vaccine and feed in a continuous process, pulse-width modulated spray nozzle control for precise vaccine application, and a programmable automation controller to regulate and monitor system performance. Initial performance evaluations demonstrated system operation within the desired design specifications. System feed rates were determined to be a function of the rotational speed (RPM) of the screw conveyor and to be linear across the operational range. Feed rates were observed to decrease by 45% over dry feed when applying liquid vaccine to the feed stream at the 220.5 ml/kg (100 ml/lb) rate, resulting in a feed rate range of 6.80-34.02 kg/min (15-75 lb/min) (95% CI). Uniform pellet-level vaccine distribution is crucial to efficacy as pellet consumption is directly correlated with fish size, with more criticality in smaller fish fed at low rates. Pellet vaccine concentrations at 6.80, 20.41, and 34.02 ml/kg were highly variable and vaccine application at all rates were observed to be statistically different (less) than the target 220.5ml/kg rate (95% CI), pointing to potential issues with vaccine delivery system configuration or inadequacies in sampling methodology. Further evaluation at the pellet level with live-attenuated vaccine to obtain viable cell counts within individual pellets would provide data necessary to address uniformity of coverage questions more fully and to develop operational protocols that maximize system capabilities and vaccine efficacy.

Os aceleradores de feixes de elétrons são utiizados em muitas aplicações, tais como pesquisa em física básica, química, medicina, biologia molecular, microeletrônica, agricultura e indústria, dentre outras. Na maioria dos aceleradores, por meio de aquecimento do filamento de tungstênio e da alta tensão via gerador Cockcroft-Walton, os elétrons passam por um campo elétrico na câmara em vácuo e são acelerados de acordo com a alta tensão aplicada, transferindo energia aos elétrons. Para fins industriais, um dos equipamentos mais utilizados é o Acelerador Industrial de Elétrons Dynamitron®. No IPEN-CNEN/ SP, instalou-se um acelerador de feixe de elétrons Dynamitron®, fabricado pela RDI Radiation Dynamics Inc., modelo DC1500/25/4, em 1978. A tecnologia analógica aplicada nesse acelerador industrial de elétrons data de 1960 e 1970. Assim, além da tecnologia de controle operacional obsoleta, não há mais peças e componentes sobressalentes originais, fornecidos pelo fabricante (RDI) ao equipamento. O objetivo deste trabalho foi desenvolvido um sistema automatizado para controle operacional do acelerador industrial de elétrons Cockcroft-Walton de 37,5 kW (1,5 MeV e 25 mA) no Instituto. / Electron beam accelerators are used in many applications, such as basic research in physics, chemistry, medicine, molecular biology, microelectronics, agriculture and industry, among others. In most accelerators, by means of heating a tungsten filament and high-voltage from a CockcroftWalton generators, the electrons go through an electrical field in the vacuum chamber and are accelerated according to the high-voltage applied to transferring energy to the electrons. For industrial purposes, one of the most used of equipment is the Dynamitron® Industrial Electron Beam Accelerator. In IPEN-CNEN/SP, this Dynamitron® Accelerator, manufactured by RDI - Radiation Dynamics Inc, model DC1500/25/4, was installed, in 1978. The analogical technology applied in this industrial accelerator dates from 1960 and 1970. Therefore, besides the obsolete operational control technology, there are not spare parts or components supplied by the manufacturer (RDI).The goal of this work was to develop an automated system for the operational control of the Cockroft-Walton industrial electrons accelerator of 37.5 kW (1.5 MeV and 25 mA), at IPEN-CNEN/SP.

Made available in DSpace on 2015-07-22T20:50:19Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-09-28 / A familia Enterobacteriaceae e uma importante causa de infeccao da corrente sanguinea (ICS). O aumento da resistencia microbiana e um fator complicador no tratamento dessas infeccoes, fazendo com que a concentracao inibitoria minima (CIM) tenha um papel importante na analise do antibiograma. No Hospital Sao Paulo - UNIFESP, ate agosto de 2007, os testes de susceptibilidade antimicrobiana (TSA) de hemoculturas para enterobacterias eram realizados atraves do metodo de discodifusao. A partir desta data, os TSA passaram a ser realizados atraves do sistema microbiologico automatizado Phoenix., que fornece uma CIM aproximada por microdiluicao. Embora a importancia da CIM seja bem documentada em estudos com Staphylococcus aureus meticilino-resistente (MRSA), nao ha estudo analisando o impacto que esses sistemas automatizados teriam sobre a evolucao clinica dessas infeccoes. Objetivos: comparar a evolucao clinica das ICS hospitalares por enterobacterias no periodo nao-automatizado versus periodo automatizado, determinar a prevalencia de enterobacterias causadoras de ICS no periodo de estudo e analisar os fatores associados a mortalidade em 14 dias das ICS hospitalares por enterobacterias. Metodo: trata-se de um estudo tipo coorte retrospectivo, onde foram avaliadas as ICS hospitalares por enterobacterias nos periodos nao-automatizado (agosto de 2006 a julho de 2007) e automatizado (agosto de 2008 a julho de 2009) do Hospital Sao Paulo - UNIFESP. Na segunda etapa do estudo, os pacientes foram realocados e divididos em obitos e sobreviventes ate 14 dias do diagnostico da infeccao para analisar os preditores de mortalidade dessas infeccoes. Resultados: foram avaliados 90 e 106 pacientes nos periodos nao-automatizado e automatizado, respectivamente. Houve a prevalencia do sexo masculino e a media de idade de 60 anos em ambos periodos. No periodo com automacao, os pacientes apresentaram maior escore APACHE II (p< 0,001), fizeram mais uso de imunossupressores (p < 0,001) e ocorreu maior incidencia de ICS associada a cateter venoso central (p = 0,002). As enterobacterias mais prevalentes foram a Klebsiella spp (36,6% e 43,3%) e Proteus spp (13,3% e 18,8%), respectivamente, no periodo nao-automatizado e automatizado. Nao houve diferenca significativa quanto a adequacao do tratamento (p = 0,45), resposta clinica (p= 0,75) e obito . 7 dias (p= 0,79), . 14 dias (p= 0,94) e . 28 dias (p= 0,12) entre periodos. Quando realocamos os pacientes que apresentaram obito ate 14 dias do diagnostico da infeccao (n= 46) em comparacao com sobreviventes em 14 dias (n= 150), os preditores que se associaram independentemente com a mortalidade foram: choque séptico (OR = 7,04; IC 95%= 2,92 – 17,01; p < 0,001) e uso de imunossupressores (OR= 2,23; IC 95%= 1,01 – 4,93; p= 0,01). Conclusão: A introdução do sistema microbiológico automatizado Phoenix™ não demonstrou impacto na evolução clínica das infecções de corrente sanguínea hospitalares por enterobactérias. Klebsiella spp e Proteus spp foram os agentes etiológicos mais prevalentes em ambos os períodos. Choque séptico e uso de imunossupressores foram os únicos fatores associados a mortalidade em 14 dias das ICS hospitalares por enterobactérias. / The family Enterobacteriaceae is an important cause of bloodstream infections (BSI). The increasing antimicrobial resistance is a complicating factor in treating these infections, thus the minimum inhibitory concentration (MIC) plays a key role in the analysis of susceptibility. Until August 2007, the antimicrobial susceptibility testing (AST) of blood cultures for Enterobacteriaceae were performed by disk diffusion method at Hospital Sao Paulo - UNIFESP. As of this date, the AST began to be performed through the Phoenix. automated microbiology system, which provides an approximated MIC by microdilution. Although the importance of MIC is well documented in studies with methicillin-resistant Staphylococcus aureus (MRSA), there is no study analyzing the impact of automated systems on clinical outcomes of these infections. Objectives: To compare the clinical outcomes of nosocomial BSI caused by Enterobacteriaceae during the non-automated versus automated period; to determine the prevalence of Enterobacteriaceae species causing BSI during the study period and analyze the factors associated with 14-day mortality. Method: This is a retrospective cohort study that evaluated nosocomial BSIs caused by Enterobacteriaceae in the periods non-automated (August 2006 to July 2007) and automated (August 2008 to July 2009) at Hospital Sao Paulo - UNIFESP. In the second part of the study, patients were realocated and divided into deaths and survivors within 14 days of diagnosis of infection to analyze the predictors of mortality. Results: We evaluated 90 and 106 patients in non-automated and automated periods, respectively. There was a male prevalence and the average age was 60 years in both periods. During the automated period, patients had higher APACHE II score (p <0.001), higher use of immunosuppressive drugs (p <0.001) and there was a greater incidence of central venous catheter-related BSI (p = 0.002). Klebsiella spp (36.6% and 43.3%) and Proteus spp (13.3% and 18.8%) respectively, during the non-automated and automated period were the most prevalent Enterobacteriaceae species. There was no significant difference regarding the adequacy of treatment (p = 0.45), clinical response (p = 0.75) and death . 7 days (p = 0.79), . 14 days (p = 0.94) and . 28 days (p = 0.12) between the periods. The predictors independently associated with mortality were septic shock (OR = 7.04; IC 95%= 2,92 . 17,01; p <0.001) and use of immunosuppressive therapy (OR = 2.23; IC 95%= 1,01 . 4,93; p = 0.01). Conclusion: The introduction of Phoenix. automated microbiology system showed no impact on clinical outcomes of bloodstream infections caused by Enterobacteriaceae. Klebsiella spp and Proteus spp were the most prevalent pathogens in both periods. Septic shock and the use of immunosuppressive drugs were the only factors associated with 14-day mortality among those patients. / TEDE / BV UNIFESP: Teses e dissertações

Made available in DSpace on 2015-05-08T14:59:51Z (GMT). No. of bitstreams: 1 parte1.pdf: 3400492 bytes, checksum: 7beeb8e70cf4eabbbfd2d39cc29c4b97 (MD5) Previous issue date: 2009-12-04 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This work presents the development of an automated system for performance analysis of solar collector plan. Concerns to the design and implementation of units of heating and cooling liquid, which provide the adjustments of circulating liquid temperature. Describes how the mass flow rate of the system is controlled by a frequency converter and pulleys that operate on a motor-pump coupling. Shows a graphical interface designed to indicate in real time: the values of the temperature of fluid flowing through the mass flow rate, the ambient air temperature, the state of solid state relays and values of solar radiation incident on solar collector. Explains how the solid state relays enable the units of heating and cooling liquid. Describes the design of electronic circuits for signal conditioning, which are responsible for the compatibility of the signals from the sensors of the system modules for A/D and D/A converters with a resolution of 12 bits. Describes how the values sampled by the A/D and D/A converters were recorded electronically and used in calculations for determining the performance of solar collectors, according to ABNT NBR-10184/1988. To confirm the functionality of the system, was evaluated a solar collector and the parameters of the resulting characteristic curve were 0.61 for FR(τα) and 6.64 W/m2 °C for FR(UL), indicating a typical solar collector behavior with aluminium absorption surface and glass coverage, traded in the Brazilian market. / Este trabalho apresenta o desenvolvimento de um sistema automatizado de levantamento de desempenho de coletor solar plano. Trata do projeto e implementação das unidades de refrigeração e de aquecimento auxiliar de líquido, as quais proporcionam os ajustes dos valores de temperatura do líquido circulante. Descreve como a vazão mássica do sistema é controlada por um conversor de freqüência e polias que atuam sobre um conjunto motorbomba. Mostra uma interface gráfica desenvolvida para indicar em tempo real: os valores da temperatura do fluido em circulação, a vazão mássica, a temperatura do ar ambiente, o estado dos atuadores acionados por relés de estado sólido e os valores da radiação solar incidente no coletor solar. Explica como os relés de estado sólido habilitam as unidades de aquecimento e resfriamento de líquido. Faz a descrição do projeto dos circuitos eletrônicos de condicionamento de sinais que são os responsáveis pela compatibilidade dos sinais oriundos dos sensores do sistema para os módulos de conversores A/D e D/A com resolução de 12 bits. Descreve como os valores amostrados pelos conversores A/D e D/A são registrados eletronicamente e usados nos cálculos para determinação do desempenho de coletores solares, de acordo com a NBR-10184/1988. Avaliou-se um coletor solar plano para confirmação da funcionalidade do sistema, cujos parâmetros da curva característica resultante foram 0,61 para FR(τα) e 6,64 W/m2°C para FR(UL), indicando um comportamento comum a coletores solares com superfície de absorção de alumínio e cobertura de vidro, comercializados no mercado brasileiro.

Актуальность темы обусловлена тем, что сегодня детским интернет-магазинам для сохранения своей конкурентоспособности на рынке детского ритейла, необходимо анализировать бизнес-процессы, устранять их недостатки, искать пути оптимизации и совершенствования бизнес-процессов. Целью данной выпускной квалификационной работы является автоматизация аналитической деятельности интернет-магазина Momchecked с применением методов машинного обучения. Практическая значимость исследования заключается в практическом применении предлагаемого способа на предприятии и получении экономической выгоды от результата внедрения информационной системы. Ключевой новизной диссертации является рассмотрение инновационных методов автоматизации аналитической отчетности предприятий, позволяющие повысить скорость обработки информации и снизить издержки компании за счет высвобождения рабочего времени сотрудников. / The relevance of the topic is due to the fact that today, for children's online stores to maintain their competitiveness in the children's retail market, it is necessary to analyze business processes, eliminate their shortcomings, and look for ways to optimize and improve business processes. The aim of this final qualification work is to automate the analytical activities of the Momchecked online store using machine learning methods. The practical significance of the study lies in the practical application of the proposed method at the enterprise and obtaining economic benefits from the result of the implementation of the information system. The key novelty of the dissertation is the consideration of innovative methods for automating the analytical reporting of enterprises, which can increase the speed of information processing and reduce company costs by freeing up staff time.