No description available.

No description available.

This study aimed to investigate the impact of uncertainties in treatment planning related to the setup and range uncertainties. Three locations of targets were chosen, one in the head, one in the abdomen, and one in the pelvis. The treatment planning was assembled by utilising a CT image of a full body anthropomorphic phantom that was implemented into the treatment planning system RayStation (version 10, RaySearch Laboratories AB). A total of five regions of interest for the target were delineated as well as the associated organs at risks. The initial treatment plan considered the objective functions that aimed to achieve coverage of the CTV according to the clinical goal whilst minimising the dose reviewed by the organs at risk. In order to take the uncertainties into account for the treatment plan robustness optimization was applied with a predetermined set of setup and range uncertainties for each target. Dose histograms for the different cases of setup and range uncertainties were used to evaluate the significance of setup and range uncertainties onto the targets and the organs at risk. In addition, dose volume histograms and the difference in dose distribution aided to visualise the greatest deviation between the different cases. For the targets located in the head the small structures and the tissue heterogeneity resulted in being more affected by the range uncertainty rather than the setup uncertainty. The targets located in the abdomen and pelvic were on the other hand more affected by the setup uncertainties. This was a result related to structures adjacent to the target with different dose constraints and tissue heterogeneities. A comparison of the dose distribution showed the effect of setup uncertainties in the two aforementioned regions. For all cases the target coverage with the robust optimisation was as expected achieved. / Syftet med denna studie var att undersöka påverkan av osäkerheterna för protonernas utbredning och patienternas positionering med avseende på planeringen av en behandlingsplan. Totalt valdes tre områden som valdes att fokusera på var av två tumörer som var placerade i huvudet, en tumör som var placerad i abdomen och en tumör som var placerad i pelvis. Behandlingsplanen konstruerades med hjälp av en CT bild för en antropomorf helkropps fantom som överfördes till behandlingsteamet RayStation (version 10, RaySearch Laboratories AB). Samtliga tumörer och relevanta organ som anses vara i riskzonen för strålning fick definierade konturer. Därefter infördes funktioner för att optimera strålning till tumörer och funktioner för att begränsa strålningen till organen som var placerade i riskzonen enligt de kliniska önskemålen. För att beakta de ingående osäkerheterna i behandlingsplanen användes optimeringsalgoritmer för att säkerhetsställa att behandlingsplanen var robust gentemot ett antal förbestämda värde på de angivna osäkerheterna. Resultaten för osäkerheterna påverkas redovisades med histograms som påvisade den erhållna dosen för tumörer respektive organ i riskzonen. Utöver histogrammen redovisades differens mellan olika tillstånd av osäkerheterna med avseende på fördelning av dos. Tumörer som var placerade i huvudet hade större påverkan av protonernas utbredning på grund av de strukturer som är små till storlek relativt tumören, men även på grund av den heterogena omgivningen i det området. Däremot var tumörer som var placerade i abdomen och i pelvis mer påverkade av patientens position. Detta på grund av att närliggande organ och strukturer hade en diskrepans i de begränsade dosnivåerna samt på grund av heterogena omgivningar. Dessa fenomen påvisades även i de jämförelser som gjordes med avseende på fördelning av dos mellan olika tillstånd av osäkerheter. Slutligen visade denna studie att robust optimering gav samtliga tumörer en dosfördelning, vilket stämmer överens med vad som var förväntat.

Made available in DSpace on 2014-06-12T17:40:08Z (GMT). No. of bitstreams: 2 arquivo605_1.pdf: 2983761 bytes, checksum: 7257ca8228b2ea15ead49ca58d51e15d (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2009 / O grande desafio das empresas tem sido crescer com um mínimo de investimentos, principalmente em um ambiente desfavorável, quando há crises e turbulências econômicas, e também de buscar melhores desempenhos a partir de métodos que as levem a conquistar e garantir produtividade e flexibilidade nos processos industriais de maneira eficiente e eficaz com agilidade e simplicidade. A metodologia SMED (Single Minute Exchange of Dies) é o principal e mais difundido conjunto de técnicas que objetiva reduzir o tempo de setup de equipamentos, minimizando tempos não produtivos e desperdícios no chão de fábrica. Através da redução dos tempos de setup, é possível melhorar o atendimento ao cliente, a produtividade, a agilidade, os níveis de desperdícios e a qualidade dos seus produtos. Este trabalho insere-se neste campo de estudo, uma vez que se propõe aplicar um método de redução dos tempos de setup, adequado-o às particularidades de uma indústria de alimentos que utiliza como sistema de gestão o gerenciamento pelo Controle da Qualidade Total, TQC (Total Quality Control Gerenciamento da Qualidade Total), de forma a aumentar a capacidade produtiva, a flexibilidade e eficiência da linha, produzindo lotes cada vez menores, e assim melhorando sua competitividade. A aplicação dessa metodologia foi precedida por uma discussão das ferramentas integradas na abordagem estudada - a clássica Troca Rápida de Ferramentas de Shigeo Shingo, a programação e o balanceamento de recursos - e aspectos concernentes ao atual perfil da indústria de alimentos brasileira, relacionando-os à importância da melhoria das atividades de setup. Também se realizou uma pesquisa bibliográfica envolvendo os conceitos do Sistema de Produção Enxuta, já que foi neste contexto que a TRF foi criada e é o sistema de gestão de operações no qual melhor se insere a TRF e ainda os conceitos do gerenciamento através do TQC, uma vez que será neste novo ambiente que a TRF será aplicada. As conclusões deste trabalho evidenciam as contribuições provenientes da aplicação da metodologia proposta, através da evolução do indicador de desempenho tempo de setup, resultando num aumento significativo de eficiência, produtividade, flexibilidade e competitividade

Estágio realizado na empresa Swedwood Portugal e orientado pelo Engº. Luís Vasques / Tese de mestrado integrado. Engenharia industrial e Gestão. Faculdade de Engenharia. Universidade do Porto. 2010

El proyecto tiene por objetivo reducir el costo de Setup en un proceso de extrusión por soplado de una empresa dedicada a la fabricación de lámina termocontraible, enfocándose en la problemática del proceso de desarrollo de muestras. Dicha problemática radica en la baja efectividad de este proceso, el cual es clave en el desarrollo de las operaciones de la organización porque de no ser satisfactorio no se concretan las ventas y se genera insatisfacción en el cliente. Además, acarrea sobrecostos como parte del Setup de producción, ya que mientras este proceso no arroje resultados positivos, la producción no iniciará y se empleará tiempo que bien se puede utilizar para producción de otros articulos. Esta falta de efectividad se produce por diversos motivos entre los que se destaca el pobre manejo del “know how”, de manera que constantemente se tiene que repetir el proceso hasta conseguir resultados. En ese sentido, se plantea la implementación de un sistema de Gestión del Conocimiento en este proceso, el cual organiza y transforma la información para generar conocimiento mediante el aprendizaje y tiene la ventaja de integrar el conocimiento implícito del trabajador, producto de años de experiencia, y transformarlo en explicito para toda la organización. Por ello, la propuesta busca reducir el tiempo de Setup al menos en un 70.31%, reduciendo en 487.71 horas el Setup anual, lo cual ahorraría 20367.81 soles anuales en costo de Setup. / The project aims to reduce the cost of Setup in a blow extrusion process of a company dedicated to the manufacture of heat shrink sheet, focusing on the problem of the sample development process. This problem lies in the low effectiveness of this process, which is key in the development of the operations of the organization because if not satisfactory, sales are not realized and customer dissatisfaction is generated. In addition, it costs overruns as part of the production setup, as long as this process does not yield positive results, production will not start and time will be used that can well be used for the production of other items. This lack of effectiveness occurs for a number of reasons, including poor management of the know-how, so that the process has to be repeated until results are achieved. In this sense, the implementation of a Knowledge Management system in this process is proposed, which organizes and transforms information to generate knowledge through learning and has the advantage of integrating the implicit knowledge of the worker , product of years of experience, and transform it into explicit for the whole organization. Therefore, the proposal seeks to reduce Setup time by at least 70.31%, reducing the annual Setup by 487.71 hours, which would save 20367.81 annual suns in Setup cost. / Tesis

<p>Exploitation of wireless sensor networks (WSNs) in ubiquitous computing environments is continuously increasing for gathering data. Contemporary distributed software systems on WSNs for pragmatic business applications have become extremely adaptive, dynamic, heterogeneous and large scaled. Management of such system is not trivial to fulfil these features, leading to more and more complex management and configuration. Along with encompassing state of art and novel techniques for such diversely dynamic system, in this thesis two alternative techniques namely “task initiation by command” and “run-time task deployment and processing” are compared, for such system’s setup and configuration. Both techniques have their own pros and cons which makes them suitable according to the requirements and contextual situations. A lot of effort has been put to make WSNs more and more efficient in terms of computations and power consumption. Hence comparative analysis of both techniques used in this report to setup and configure WSN can be a benchmark to lead towards most appropriate solution to compensate the need of efficient energy and resource consumption.Both alternative schemes are implemented to setup WSN on Sun Microsystems sunSPOT (Small Programmable Object Technology) sensor nodes which are embedded microcontrollers and programmed them in java (j2me). It performs radio communication between wireless sensors and host via sink node also called base station, along with over the air run-time management of sensors. SunSPOTs built in libraries and KSN libraries are used to implement these alternatives and compare the memory footprint, communication pattern and energy consumption.Exploitation of wireless sensor networks (WSNs) in ubiquitous computing environments is continuously increasing for gathering data. Contemporary distributed software systems on WSNs for pragmatic business applications have become extremely adaptive, dynamic, heterogeneous and large scaled. Management of such system is not trivial to fulfil these features, leading to more and more complex management and configuration. Along with encompassing state of art and novel techniques for such diversely dynamic system, in this thesis two alternative techniques namely “task initiation by command” and “run-time task deployment and processing” are compared, for such system’s setup and configuration. Both techniques have their own pros and cons which makes them suitable according to the requirements and contextual situations. A lot of effort has been put to make WSNs more and more efficient in terms of computations and power consumption. Hence comparative analysis of both techniques used in this report to setup and configure WSN can be a benchmark to lead towards most appropriate solution to compensate the need of efficient energy and resource consumption.Both alternative schemes are implemented to setup WSN on Sun Microsystems sunSPOT (Small Programmable Object Technology) sensor nodes which are embedded microcontrollers and programmed them in java (j2me). It performs radio communication between wireless sensors and host via sink node also called base station, along with over the air run-time management of sensors. SunSPOTs built in libraries and KSN libraries are used to implement these alternatives and compare the memory footprint, communication pattern and energy consumption.</p>

It is nice to give animators artistic freedom, but having to animate every bounce, sway, and counter-balancing movement of a rider on a horse isn't freedom at all. It is painstaking labor that could easily be prevented with an effective character setup. If an animation piece is only going to have a few shots with a horse and rider, then the trouble of setting up an automated character rig is not practical, but if there are a significant amount of shots with a horse and rider galloping across the prairie, doing death defying stunts, and walking for an extended time into the sunset then there needs to be a way to automate the reactions of the rider to the horse. This thesis focuses on what parts of a horse one can analyze to know at what point a rider will lean forward, bounce up from the saddle, or in any way react to a variety of different horse movements. The automated character setup, or rig, makes animating a rider on a horse much more efficient.

Exploitation of wireless sensor networks (WSNs) in ubiquitous computing environments is continuously increasing for gathering data. Contemporary distributed software systems on WSNs for pragmatic business applications have become extremely adaptive, dynamic, heterogeneous and large scaled. Management of such system is not trivial to fulfil these features, leading to more and more complex management and configuration. Along with encompassing state of art and novel techniques for such diversely dynamic system, in this thesis two alternative techniques namely “task initiation by command” and “run-time task deployment and processing” are compared, for such system’s setup and configuration. Both techniques have their own pros and cons which makes them suitable according to the requirements and contextual situations. A lot of effort has been put to make WSNs more and more efficient in terms of computations and power consumption. Hence comparative analysis of both techniques used in this report to setup and configure WSN can be a benchmark to lead towards most appropriate solution to compensate the need of efficient energy and resource consumption.Both alternative schemes are implemented to setup WSN on Sun Microsystems sunSPOT (Small Programmable Object Technology) sensor nodes which are embedded microcontrollers and programmed them in java (j2me). It performs radio communication between wireless sensors and host via sink node also called base station, along with over the air run-time management of sensors. SunSPOTs built in libraries and KSN libraries are used to implement these alternatives and compare the memory footprint, communication pattern and energy consumption.Exploitation of wireless sensor networks (WSNs) in ubiquitous computing environments is continuously increasing for gathering data. Contemporary distributed software systems on WSNs for pragmatic business applications have become extremely adaptive, dynamic, heterogeneous and large scaled. Management of such system is not trivial to fulfil these features, leading to more and more complex management and configuration. Along with encompassing state of art and novel techniques for such diversely dynamic system, in this thesis two alternative techniques namely “task initiation by command” and “run-time task deployment and processing” are compared, for such system’s setup and configuration. Both techniques have their own pros and cons which makes them suitable according to the requirements and contextual situations. A lot of effort has been put to make WSNs more and more efficient in terms of computations and power consumption. Hence comparative analysis of both techniques used in this report to setup and configure WSN can be a benchmark to lead towards most appropriate solution to compensate the need of efficient energy and resource consumption.Both alternative schemes are implemented to setup WSN on Sun Microsystems sunSPOT (Small Programmable Object Technology) sensor nodes which are embedded microcontrollers and programmed them in java (j2me). It performs radio communication between wireless sensors and host via sink node also called base station, along with over the air run-time management of sensors. SunSPOTs built in libraries and KSN libraries are used to implement these alternatives and compare the memory footprint, communication pattern and energy consumption.

With recent growth in mobile components and advances in wireless communication technology, mobile computing is an increasingly important area of research. Enabling mobility in IP networks becomes more and more significant. Mobile IP is the only current means for offering seamless roaming to mobile computers in the Internet. It has recently progressed along the ladder to standardization within the Internet Engineering Task Force (IETF). However, Mobile IP suffers from so-called triangular-routing problem and packet loss due to handoff. In this paper, we propose a new architecture to solve these two problems noted above. We assume that network routers such as Home Agent¡BForeign Agent¡BGeneral Router can equip our Pre-Accessing Engine (Mobile IP Engine) to access IP packets before IP routing Engine. With our Mobile IP table established in Mobile IP Engine, we can set up a transmission path for each correspondent node (CN) to transmit data to mobile node (MN). Therefore, we not only solve triangular-routing problem by direct transmission path, but also decrease handoff latency from path updating function. Furthermore, we use retransmission and flush schemes to avoid packet loss and out-of-order problems. Finally, we use simulation to prove our architecture is useful.