Redes definidas por software (Software-Defined Networking – SDN) tem como objetivo fornecer uma arquitetura mais sofisticada e precisa para gerenciar e monitorar o tráfego da rede. SDN permite centralizar parte da lógica de tomada de decisão sobre o processamento de fluxo e roteamento de pacotes em dispositivos chamados controladores. Apesar disso, o comportamento dos dispositivos de rede e suas configurações são muitas vezes escritos para situações específicas diretamente no controlador. Isto torna-se um problema quando há um aumento no número de elementos, ligações e serviços de rede, resultando numa grande quantidade de regras e uma elevada sobrecarga relacionada à configuração da rede. Como alternativa , técnicas, tais como gerenciamento baseado em políticas (Policy-Based Management – PBM) e refinamento de políticas podem ser utilizadas por operadores de alto nível para escrever Acordos de Nível de Serviço (Service Level Agreements – SLAs) em uma interface amigável, sem a necessidade de alterar o código implementado nos controladores. No entanto, o refinamento de políticas na nova área de pesquisa SDN tem sido um tema negligenciado, em parte, porque o refinamento não é um processo trivial. Ao utilizar SLAs, a sua tradução para políticas de baixo nível, por exemplo, regras para a configuração de elementos de comutação, não é simples. Se essa tradução não for realizada corretamente, os elementos do sistema podem não ser capaz de cumprir os requisitos implícitos especificados no SLA. Neste contexto, este trabalho apresenta ARKHAM: um avançado conjunto de ferramentas de refinamento para manipulação de acordos de nível de serviço em redes definidas por software. Este conjunto de ferramentas é composto por (i) um framework para criação de políticas que usa raciocínio lógico para a especificação de objetivos de nível de negócio e automatização de seu refinamento; (ii) um controlador OpenFlow que realiza a coleta de informações e implantação de configurações na rede; e (iii) uma representação formal de políticas de alto nível utilizando Event Calculus e aplicando raciocínio lógico para modelar tanto o comportamento do sistema quanto o processo de refinamento de políticas para o gerenciamento de SDN. Como resultado, a abordagem é capaz de identificar as necessidades e os recursos que precisam ser configurados de acordo com o refinamento do SLA, podendo assim configurar e executar com sucesso ações dinâmicas de suporte à reconfiguração de infraestrutura. / Software-Defined Networking (SDN) aims to provide a more sophisticated and accurate architecture for managing and monitoring network traffic. SDN permits centralizing part of the decision-making logic regarding flow processing and packet routing in controller devices. Despite this, the behavior of network devices and their configurations are often written for specific situations directly in the controller. This becomes an issue when there is an increase in the number of network elements, links, and services, resulting in a large amount of rules and a high overhead related to network configuration. As an alternative, techniques such as Policy- Based Management (PBM) and policy refinement can be used by high-level operators to write Service Level Agreements (SLAs) in a user-friendly interface without the need to change the code implemented in the controllers. However, policy refinement in the new research area of SDN has been a neglected topic, in part, because refinement is a nontrivial process. When using SLAs, their translation to low-level policies, e.g., rules for configuring switching elements, is not straightforward. If this translation is not performed properly, the system elements may not be able to meet the implicit requirements specified in the SLA. In this context, we introduce ARKHAM: an Advanced Refinement Toolkit for Handling Service Level Agreements in Software-Defined Networking. This work presents (i) a Policy Authoring Framework that uses logical reasoning for the specification of business-level goals and to automate their refinement; (ii) an OpenFlow controller which performs information gathering and configuration deployment; and (iii) a formal representation using event calculus that describes our solution. As a result, our approach is capable of identifying the requirements and resources that need to be configured in accordance with SLA refinement, and can successfully configure and execute dynamic actions for supporting infrastructure reconfiguration.

Aim of this thesis is optimisation of system of managing stocks in Viscofan CZ s. r. o. and suggestion of measure to reduce lock capital in stocks with increase of customer service level. On the base of ABC analyse was chosen skinless and fibrous products. Skinless is main product, on the other hand fibrous is minor. Chosen casing was analysed to turnover of stock, lock capital and structure of stock in the respect of time of storage. On the base of these results was found out that approximately in the skinless is locked about 69 mil. Kč. In the fibrous it is more than 12 mil. Kč. Necessary to say, the outdate stocks of skinless form 18 % of whole stock that represent 12,5 mil. Kč. In case of fibrous form 46 % that mean 5,56 mil. Kč. Old stocks arise because of refusal of acceptance goods by customers. From this situation result need of optimization of system of managing stocks to lower lock capital. Two types of optimisation were suggested. First type tries to minimize amount of new outdate stock. First I suggest to start analysing of customer service level. This remedy could lead to reduce new outdate stock about 50 % that means saving almost 5,7 mil. Kč. Important benefit could be higher reliability of company for customers. With this optimization is connected another that suggest implementation of participation of sales representatives on income from sold product with reflection to reliability of their customers. The more reliable customer the higher commission. Second type of optimization is way of reduce current old stocks. One possibility is to compensate new orders by outdate stocks. This could lead to provide lock capital, mostly 20 mil. Kč. Second possibility is to sale with discount. All suggested optimizations start from current needs of company. Combination of optimization could bring higher benefits than individual implementation.

Redes definidas por software (Software-Defined Networking – SDN) tem como objetivo fornecer uma arquitetura mais sofisticada e precisa para gerenciar e monitorar o tráfego da rede. SDN permite centralizar parte da lógica de tomada de decisão sobre o processamento de fluxo e roteamento de pacotes em dispositivos chamados controladores. Apesar disso, o comportamento dos dispositivos de rede e suas configurações são muitas vezes escritos para situações específicas diretamente no controlador. Isto torna-se um problema quando há um aumento no número de elementos, ligações e serviços de rede, resultando numa grande quantidade de regras e uma elevada sobrecarga relacionada à configuração da rede. Como alternativa , técnicas, tais como gerenciamento baseado em políticas (Policy-Based Management – PBM) e refinamento de políticas podem ser utilizadas por operadores de alto nível para escrever Acordos de Nível de Serviço (Service Level Agreements – SLAs) em uma interface amigável, sem a necessidade de alterar o código implementado nos controladores. No entanto, o refinamento de políticas na nova área de pesquisa SDN tem sido um tema negligenciado, em parte, porque o refinamento não é um processo trivial. Ao utilizar SLAs, a sua tradução para políticas de baixo nível, por exemplo, regras para a configuração de elementos de comutação, não é simples. Se essa tradução não for realizada corretamente, os elementos do sistema podem não ser capaz de cumprir os requisitos implícitos especificados no SLA. Neste contexto, este trabalho apresenta ARKHAM: um avançado conjunto de ferramentas de refinamento para manipulação de acordos de nível de serviço em redes definidas por software. Este conjunto de ferramentas é composto por (i) um framework para criação de políticas que usa raciocínio lógico para a especificação de objetivos de nível de negócio e automatização de seu refinamento; (ii) um controlador OpenFlow que realiza a coleta de informações e implantação de configurações na rede; e (iii) uma representação formal de políticas de alto nível utilizando Event Calculus e aplicando raciocínio lógico para modelar tanto o comportamento do sistema quanto o processo de refinamento de políticas para o gerenciamento de SDN. Como resultado, a abordagem é capaz de identificar as necessidades e os recursos que precisam ser configurados de acordo com o refinamento do SLA, podendo assim configurar e executar com sucesso ações dinâmicas de suporte à reconfiguração de infraestrutura. / Software-Defined Networking (SDN) aims to provide a more sophisticated and accurate architecture for managing and monitoring network traffic. SDN permits centralizing part of the decision-making logic regarding flow processing and packet routing in controller devices. Despite this, the behavior of network devices and their configurations are often written for specific situations directly in the controller. This becomes an issue when there is an increase in the number of network elements, links, and services, resulting in a large amount of rules and a high overhead related to network configuration. As an alternative, techniques such as Policy- Based Management (PBM) and policy refinement can be used by high-level operators to write Service Level Agreements (SLAs) in a user-friendly interface without the need to change the code implemented in the controllers. However, policy refinement in the new research area of SDN has been a neglected topic, in part, because refinement is a nontrivial process. When using SLAs, their translation to low-level policies, e.g., rules for configuring switching elements, is not straightforward. If this translation is not performed properly, the system elements may not be able to meet the implicit requirements specified in the SLA. In this context, we introduce ARKHAM: an Advanced Refinement Toolkit for Handling Service Level Agreements in Software-Defined Networking. This work presents (i) a Policy Authoring Framework that uses logical reasoning for the specification of business-level goals and to automate their refinement; (ii) an OpenFlow controller which performs information gathering and configuration deployment; and (iii) a formal representation using event calculus that describes our solution. As a result, our approach is capable of identifying the requirements and resources that need to be configured in accordance with SLA refinement, and can successfully configure and execute dynamic actions for supporting infrastructure reconfiguration.

Redes definidas por software (Software-Defined Networking – SDN) tem como objetivo fornecer uma arquitetura mais sofisticada e precisa para gerenciar e monitorar o tráfego da rede. SDN permite centralizar parte da lógica de tomada de decisão sobre o processamento de fluxo e roteamento de pacotes em dispositivos chamados controladores. Apesar disso, o comportamento dos dispositivos de rede e suas configurações são muitas vezes escritos para situações específicas diretamente no controlador. Isto torna-se um problema quando há um aumento no número de elementos, ligações e serviços de rede, resultando numa grande quantidade de regras e uma elevada sobrecarga relacionada à configuração da rede. Como alternativa , técnicas, tais como gerenciamento baseado em políticas (Policy-Based Management – PBM) e refinamento de políticas podem ser utilizadas por operadores de alto nível para escrever Acordos de Nível de Serviço (Service Level Agreements – SLAs) em uma interface amigável, sem a necessidade de alterar o código implementado nos controladores. No entanto, o refinamento de políticas na nova área de pesquisa SDN tem sido um tema negligenciado, em parte, porque o refinamento não é um processo trivial. Ao utilizar SLAs, a sua tradução para políticas de baixo nível, por exemplo, regras para a configuração de elementos de comutação, não é simples. Se essa tradução não for realizada corretamente, os elementos do sistema podem não ser capaz de cumprir os requisitos implícitos especificados no SLA. Neste contexto, este trabalho apresenta ARKHAM: um avançado conjunto de ferramentas de refinamento para manipulação de acordos de nível de serviço em redes definidas por software. Este conjunto de ferramentas é composto por (i) um framework para criação de políticas que usa raciocínio lógico para a especificação de objetivos de nível de negócio e automatização de seu refinamento; (ii) um controlador OpenFlow que realiza a coleta de informações e implantação de configurações na rede; e (iii) uma representação formal de políticas de alto nível utilizando Event Calculus e aplicando raciocínio lógico para modelar tanto o comportamento do sistema quanto o processo de refinamento de políticas para o gerenciamento de SDN. Como resultado, a abordagem é capaz de identificar as necessidades e os recursos que precisam ser configurados de acordo com o refinamento do SLA, podendo assim configurar e executar com sucesso ações dinâmicas de suporte à reconfiguração de infraestrutura. / Software-Defined Networking (SDN) aims to provide a more sophisticated and accurate architecture for managing and monitoring network traffic. SDN permits centralizing part of the decision-making logic regarding flow processing and packet routing in controller devices. Despite this, the behavior of network devices and their configurations are often written for specific situations directly in the controller. This becomes an issue when there is an increase in the number of network elements, links, and services, resulting in a large amount of rules and a high overhead related to network configuration. As an alternative, techniques such as Policy- Based Management (PBM) and policy refinement can be used by high-level operators to write Service Level Agreements (SLAs) in a user-friendly interface without the need to change the code implemented in the controllers. However, policy refinement in the new research area of SDN has been a neglected topic, in part, because refinement is a nontrivial process. When using SLAs, their translation to low-level policies, e.g., rules for configuring switching elements, is not straightforward. If this translation is not performed properly, the system elements may not be able to meet the implicit requirements specified in the SLA. In this context, we introduce ARKHAM: an Advanced Refinement Toolkit for Handling Service Level Agreements in Software-Defined Networking. This work presents (i) a Policy Authoring Framework that uses logical reasoning for the specification of business-level goals and to automate their refinement; (ii) an OpenFlow controller which performs information gathering and configuration deployment; and (iii) a formal representation using event calculus that describes our solution. As a result, our approach is capable of identifying the requirements and resources that need to be configured in accordance with SLA refinement, and can successfully configure and execute dynamic actions for supporting infrastructure reconfiguration.

El presente estudio se realizó en la empresa Soprin SAC pertenece al rubro de pinturas, cuenta con dos líneas de producción: la línea de temple y la línea de látex. El estudio estará enfocado a la línea de látex. Actualmente, se fabrican los productos terminados (PT) de acuerdo a las cantidades especificadas en las órdenes de compra de los clientes, las cuales son ingresadas al módulo de ventas para luego ser aprobadas por el área de Créditos y Cobranzas, siendo finalmente atendidas por el área de Planeamiento y Control de la Producción que se encargará de generar las ordenes fabricación y envasado. Las órdenes de fabricación y envasado generadas por el área de planeamiento para cubrir los pedidos son lanzadas de forma inadecuada, ya que solo tienen la finalidad de satisfacer las OV’s, pero no optimizando los recursos y la capacidad de la planta. Constantemente hay falta de insumos oportunos para la fabricación y envasado; lo que genera: Capacidad de planta ociosa, Desabastecimiento de productos terminados y Bajo nivel de servicio de órdenes de venta. Aplicando las metodologías del Planeamiento de Requerimiento de Materiales, Análisis del Lote Estándar y el Análisis de la demanda; logramos incrementar nuestra Capacidad de Planta. La cantidad de litros producidos por lote es mayor; dando como resultado reducir los costos de energía, mano de obra, tiempo de fabricación, tiempo de preparación, cantidad de veces que Almacén Materias Primas nos despacha la materia prima y material de envase, cantidad de veces que se entregan los productos terminados a Almacén PT, y el tiempo de matizado de los colores. El porcentaje de nivel de servicio se incrementa en un 28.6%, tomando como punto de partida el promedio de los meses Enero-Mayo 2015 y comparándolo con el obtenido en Septiembre 2015. Esto refleja la eficiencia en la atención de las OV’s por parte del Área de Producción. This study was conducted at the company Soprin SAC belongs to the category of paints, it has two production lines: Temple line and the Latex line. The study will focus on the Latex line. Currently, the finished products (FP) according to the amounts specified in purchase orders from customers, which are entered into the sales module then be approved by the area of Credit and Collections, finally being addressed are manufactured by area Planning and Production Control which is responsible for generating the manufacturing and packaging orders. The manufacturing and packaging orders generated by the planning area to cover the orders are released improperly, as they only have the purpose of meeting the OV’s, but optimizing resources and the capacity of the plant. Constantly there is a lack of timely inputs for the manufacture and packaging; generating: idle plant capacity, shortage of finished products and services Low selling orders. Applying methodologies Material Requirement Planning, Analysis Standard Lot and Demand Analysis; we increase our plant capacity. The number of liters produced per batch is greater; resulting in lower energy costs, labor, manufacturing time, preparation time, number of times we Warehouse Raw Materials shipped raw material and packaging material, number of times to store finished products are delivered FP and time hueing of colors. The percentage of service level increases by 28.6%, taking as a starting point the average of the months January to May 2015 and compared with that obtained in September 2015. This reflects the efficiency of care by the OV's Production area.

SAs are documents that specify the business relationship between stakeholders to an outsourcing agreement. SAs specify this relationship in a legally binding manner that assists in managing expectations of the stakeholders about the service provision. According to Verma (1999), an SA is a precise statement of the expectations and obligations that exist in a business relationship between two organisation: the service provider and the client. In order for organizations to have successful outsourcing partnerships, they need well crafted methods of developing Service Agreements (SAs). Successful methods will produce a conclusive contract that will act as a working document that details the spirit of cooperation between the service provider and the service recipient. This research investigates the development of SAs in the Information and Communication Technology (ICT) sector, and proposes a model for their development. A number of models for SA development have been analysed. Models are analysed from leading researchers in the area, from software houses such as Microsoft and from international standards organisations such as the BS15000 which stipulates the ITIL framework. Eight development principles are identified and explored. An investigation into SAs and their development is conducted. A model is proposed that is composed of the development principles. The development of SAs was explored in an empirical study by means of a survey administered to industry practitioners and a series of interviews with managers in the ICT industry. The results of the study indicate varying levels of support for the development principles and limited relationship between the development principles and the success of the SA, as defined by the number of changes made to the SA after it is completed.

Management of Virtual Organizations faces new challenges that traditional approaches cannot address. This research proposes a performance measurement framework for service oriented virtual organizations including a structural and a procedural component. The structural framework aligns the activities of partners in a virtual organization at three different layers. The first layer is designed for partners’ strategic alignment through coordination of the value creation network. In the second layer, performance dimensions of partners’ collaboration are defined and mapped to the service choreography model. The third layer focuses on assessing effectiveness and efficiency of partners’ domain specific services, which is designed based on ITIL V3 service level management guidelines. In order to consolidate the structural framework, these three layers are integrated using a method for extracting service choreography model and SLA aggregation patterns from the value network. The procedural framework, on the other side, defines the processes required to design the KPI structure, implement the solution, communicate the results, and derive improvements. We propose an implementation architecture that enables inter-organizational performance management in collaborative environments. Then, the IBM products for business process and performance management (IBM BPM, Business Monitor, and Cognos BI) are employed to implement the proposed architecture. The conceptual framework along with the implementation architecture provides an integrated solution for decentralized performance measurement without the need for a central authority. We demonstrate that the proposed solution enhances flexibility, scalability, and interoperability, and supports transparency of partners’ performance information at an agreed-upon level as a basis for mutual trust.

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / The present research seeks to show the importance of applying process management techniques and food safety norms in the operational processes of the supply chain to know the meaning and the need for an integrated hybrid model. The article reviews the sugarcane distillery sector of Peru and its main operational problems. Based on the literature reviewed and discussed with academics who have knowledge of the food supply chain, an integrated hybrid model was developed to help any distillery with lower levels of competitiveness than its competitors in other sectors, such as pisco, applying techniques of process management and food safety to increase the efficiency of liquor distilleries. The findings confirm that distilleries can increase their efficiency, thanks to the higher performance of their operations after their alignment with the integrated model. / Revisión por pares

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / The present research seeks to show the importance of applying process management techniques and food safety norms in the operational processes of the supply chain to know the meaning and the need for an integrated hybrid model. The article reviews the sugarcane distillery sector of Peru and its main operational problems. Based on the literature reviewed and discussed with academics who have knowledge of the food supply chain, an integrated hybrid model was developed to help any distillery with lower levels of competitiveness than its competitors in other sectors, such as pisco, applying techniques of process management and food safety to increase the efficiency of liquor distilleries. The findings confirm that distilleries can increase their efficiency, thanks to the higher performance of their operations after their alignment with the integrated model. / Revisión por pares

[pt] A utilização do modelo para ressuprimento automático de estoque pelo fornecedor (Vendor Management Inventory) é um dos fatores importantes para as empresas promoverem um equilíbrio entre o nível de serviço acordado com o cliente e o custo logístico inerente à operação de gestão de estoques. Devido à importância do tema, e com a oportunidade da autora participar da implementação do programa VMI em uma empresa no ramo de distribuição de combustíveis para gestão de estoques de postos de combustíveis, o objetivo desta dissertação foi avaliar o quão efetivo é a metodologia VMI desenvolvida e adotada na melhoria dos indicadores de desempenho. Para tal, foram feitas análises de seis indicadores para cinco clientes envolvidos nesta operação e, realizada a comparação entre os programas VMI em duas empresas distintas, buscando avaliar os fatores críticos, as melhores práticas e vantagens competitivas. A metodologia utilizada para a elaboração desta dissertação contempla pesquisas bibliográficas, dados de fontes primárias extraídos de sistemas de informação da empresa, entrevistas com funcionários envolvidos neste processo, e visitas in loco para observação direta. Os resultados desta dissertação foram a análise da efetividade do programa VMI para a melhoria dos indicadores de desempenho, os motivos que contribuíram para estes resultados e a proposta de melhoria do programa VMI baseada na comparação da aplicação do programa em empresas no mesmo segmento. / [en] The use of an efficient model for inventory replenishment by the supplier, VMI (Vendor Management Inventory) is one of the most important factors for companies to promote a balanced trade-off between the service level offered to the customer and the logistics costs inherent in the operation of inventory management. Due to the importance of this subject, and the opportunity of the author to take part in the implementation of the VMI program in a company of the fuel distribution sector for inventory management of gas stations, the objective of this master’s thesis is to evaluate how effective the VMI methodology is to improve performance indicators. To this end, the six indicators for five customers involved in this operation were analyzed and a comparison made among the VMI programs between two different companies, to evaluate the critical success factors, best practices and competitive advantages. The methodology used to prepare this dissertation includes bibliographic research, extracted data from primary sources of information system of the company, interviews with employees involved in this process, as well as in situ visits for direct observation. The results of this thesis were to analysis the effectiveness of the VMI program for the improvement of the performance indicators, the reasons that contributed to these results and the proposal of improvement actions for the VMI program based on the comparison of the program in two different companies of the same segment.