Em diversos setores, em especial no automotivo, uma boa gestão dos estoques de peças de reposição tem impacto significativo na satisfação dos clientes e em sua fidelidade aos fabricantes. Neste trabalho foram estudadas diferentes políticas de gestão dos estoques de peças de reposição, para comparar seu desempenho e elaborar recomendações para seu uso. Foram comparados 17 conjuntos de políticas que envolvem diferentes abordagens no registro das demandas (dados individuais de cada pedido versus dados agregados em janelas de tempo semanais e mensais), modelos de previsão (média móvel, Croston modificado SBA) e diferentes formas de modelar a distribuição da demanda durante o Lead-time de ressuprimento (Normal, Gama, Binomial Negativa, composta Poisson-Normal, composta Poisson-Gama). Cada um desses 17 conjuntos de políticas foi simulado sob duas dinâmicas de reparametrização (mensal e semestral) e para quatro objetivos diferentes do nível de serviço (TFR: Target Fill Rate), totalizando 136 simulações para cada item do estoque (SKU). Foram considerados 10.032 SKU\'s de uma montadora de automóveis instalada no Brasil, com um histórico de seis anos de movimentação. Diferentes recomendações foram elaboradas conforme categorização dos itens já existente na literatura. Os resultados apontaram recomendações distintas para cada TFR, incluindo combinações de todas as alternativas estudadas, descartando apenas o uso das distribuições Normal, composta Poisson-Normal e composta Poisson-Gama. Sugere-se que as recomendações sirvam como guia para o uso desses modelos pelos praticantes. / In many areas, including automotive, a good spare parts inventory management can substantially affect customer satisfaction and their loyalty to the brands. Different spare parts inventory control policies were evaluated in this study, aiming to compare their performance and write recommendations for their usage. Seventeen policy sets were compared, including different approaches in recording demand data (individual orders data against time bucket records weekly and monthly), different demand forecasting methods (simple moving average, Syntetos-Boylan-approximation SBA) and different models for demand distribution during lead-time (Normal, Gama, Negative Binomial, compound Poisson-Normal, compound Poisson-Gama). Each policy set was simulated under two revision frequencies (monthly and semi-annually) and four different Target-Fill-Rates (TFR), totalizing 136 simulation runs for each SKU. Database included movement of 10.032 SKU´s during last 6 years from an automaker installed in Brazil. Results pointed different recommendations for existing classification schemes and under each TFR. Recommendations included all studied alternatives, discarding only the usage of Normal, compound Poisson-Normal and compound Poisson-Gama for demand distribution during lead-time. Practitioners are stimulated to use these recommendations as a guideline.

Atualmente, as empresas buscam garantir disponibilidade de produto ao cliente final, com o menor nível de inventário possível. Isto ocorre devido a diversidade crescente no número de produtos e o elevado custo de oportunidade do capital. As ferramentas de simulação disponibilizam aos gestores melhores visões do negócio e melhores condições para a tomada de decisão. Em um Centro de Distribuição de Peças (CDP), para se alcançar um melhor desempenho no gerenciamento do inventário é importante poder contar com uma ferramenta capacitada para realizações de simulações de possíveis cenários. Este trabalho tem como objetivo desenvolver um ambiente de simulação do inventário de um CDP. O ambiente de simulação permite realizar simulações visando obter como resultados a previsão de demanda, a política de reposição das peças, o percentual de atendimento dos pedidos, os meses de inventário, a quantidade de scrap e a quantidade de peças para devolução. Para o desenvolvimento do ambiente de simulação utilizou-se de métodos de modelagem e linguagem de programação orientado a objeto. O ambiente de simulação foi submetido a experimentos com três cenários diferentes e os resultados confirmaram a qualidade do sistema de simulação proposto. / Nowadays companies are trying to become their product available to final client with the smallest possible level of inventory. The simulation tools offer to managers best views of business and best condition to decide. In a Part Distribution Center (PDC), to reach the best performance in the management of the inventory it is important to be able to count on a tool that is capable of accomplishments from simulation possible sceneries. The objective of this work is to develop a simulation environment of the inventory of a PDC. The simulation environment performs projections in order to get demand forecasts, the parts replenishment policy, the attendance percentage, the inventory months, scrap quantity and quantity part to devolution. To develop the simulation environment are used tools like modeling methods and programming object-oriented language. The simulation environment was submitted to experiments with three different sceneries and the results confirmed the quality of the simulation system proposed.

Em diversos setores, em especial no automotivo, uma boa gestão dos estoques de peças de reposição tem impacto significativo na satisfação dos clientes e em sua fidelidade aos fabricantes. Neste trabalho foram estudadas diferentes políticas de gestão dos estoques de peças de reposição, para comparar seu desempenho e elaborar recomendações para seu uso. Foram comparados 17 conjuntos de políticas que envolvem diferentes abordagens no registro das demandas (dados individuais de cada pedido versus dados agregados em janelas de tempo semanais e mensais), modelos de previsão (média móvel, Croston modificado SBA) e diferentes formas de modelar a distribuição da demanda durante o Lead-time de ressuprimento (Normal, Gama, Binomial Negativa, composta Poisson-Normal, composta Poisson-Gama). Cada um desses 17 conjuntos de políticas foi simulado sob duas dinâmicas de reparametrização (mensal e semestral) e para quatro objetivos diferentes do nível de serviço (TFR: Target Fill Rate), totalizando 136 simulações para cada item do estoque (SKU). Foram considerados 10.032 SKU\'s de uma montadora de automóveis instalada no Brasil, com um histórico de seis anos de movimentação. Diferentes recomendações foram elaboradas conforme categorização dos itens já existente na literatura. Os resultados apontaram recomendações distintas para cada TFR, incluindo combinações de todas as alternativas estudadas, descartando apenas o uso das distribuições Normal, composta Poisson-Normal e composta Poisson-Gama. Sugere-se que as recomendações sirvam como guia para o uso desses modelos pelos praticantes. / In many areas, including automotive, a good spare parts inventory management can substantially affect customer satisfaction and their loyalty to the brands. Different spare parts inventory control policies were evaluated in this study, aiming to compare their performance and write recommendations for their usage. Seventeen policy sets were compared, including different approaches in recording demand data (individual orders data against time bucket records weekly and monthly), different demand forecasting methods (simple moving average, Syntetos-Boylan-approximation SBA) and different models for demand distribution during lead-time (Normal, Gama, Negative Binomial, compound Poisson-Normal, compound Poisson-Gama). Each policy set was simulated under two revision frequencies (monthly and semi-annually) and four different Target-Fill-Rates (TFR), totalizing 136 simulation runs for each SKU. Database included movement of 10.032 SKU´s during last 6 years from an automaker installed in Brazil. Results pointed different recommendations for existing classification schemes and under each TFR. Recommendations included all studied alternatives, discarding only the usage of Normal, compound Poisson-Normal and compound Poisson-Gama for demand distribution during lead-time. Practitioners are stimulated to use these recommendations as a guideline.

Atualmente, as empresas buscam garantir disponibilidade de produto ao cliente final, com o menor nível de inventário possível. Isto ocorre devido a diversidade crescente no número de produtos e o elevado custo de oportunidade do capital. As ferramentas de simulação disponibilizam aos gestores melhores visões do negócio e melhores condições para a tomada de decisão. Em um Centro de Distribuição de Peças (CDP), para se alcançar um melhor desempenho no gerenciamento do inventário é importante poder contar com uma ferramenta capacitada para realizações de simulações de possíveis cenários. Este trabalho tem como objetivo desenvolver um ambiente de simulação do inventário de um CDP. O ambiente de simulação permite realizar simulações visando obter como resultados a previsão de demanda, a política de reposição das peças, o percentual de atendimento dos pedidos, os meses de inventário, a quantidade de scrap e a quantidade de peças para devolução. Para o desenvolvimento do ambiente de simulação utilizou-se de métodos de modelagem e linguagem de programação orientado a objeto. O ambiente de simulação foi submetido a experimentos com três cenários diferentes e os resultados confirmaram a qualidade do sistema de simulação proposto. / Nowadays companies are trying to become their product available to final client with the smallest possible level of inventory. The simulation tools offer to managers best views of business and best condition to decide. In a Part Distribution Center (PDC), to reach the best performance in the management of the inventory it is important to be able to count on a tool that is capable of accomplishments from simulation possible sceneries. The objective of this work is to develop a simulation environment of the inventory of a PDC. The simulation environment performs projections in order to get demand forecasts, the parts replenishment policy, the attendance percentage, the inventory months, scrap quantity and quantity part to devolution. To develop the simulation environment are used tools like modeling methods and programming object-oriented language. The simulation environment was submitted to experiments with three different sceneries and the results confirmed the quality of the simulation system proposed.

The purpose of this thesis was to automate verification of the software requirements for an implantable cardioverter defibrillator with minimal manual rework. The requirements were written in plain English with only loose stylistic constraints. While full automation proved infeasible, many significant advances were made towards solving the problem, including a framework for storing requirements, a program which translates most of the natural language requirements into the framework, and a novel approach to parts of speech analysis.

Reuse of components as opposed to material recovery, recycling or disposal has been identified as one of the most efficient EOL strategies for products. The concept behind reuse is that some components and subassemblies have a design life that exceeds the life of the product itself. In order for reuse to be successfully implemented as an EOL strategy, a designer needs to incorporate into a product a philosophy of Design for Reuse (DfRe) at the early design stage. Reliable methods to assess the remaining life of used components based on a products usage life are also required. Furthermore, current industry practices and literature advocate that there is no methodology to decide which parameters need to be redesigned so as to change the life of a selected component to a desired level. The objective of this research is to develop a methodology to assess the reuse potential of product groups based on component failure mechanisms and their associated critical lifetime prediction design parameters. Utilising these clustered groups mathematical models were then developed to establish the useful life of the components for each clustered group. Finally, a means of equating useful life to design life was established and the relationship between, the failure mechanisms, critical lifetime prediction design parameters and design life were represented in graphical format. In order to achieve the proposed objective, Cluster analysis, in particular Group Technology (GT) and Hierarchical clustering were employed to group components with similar failure mechanisms. Following this, multiple linear regression was used to establish mathematical models based on condition monitoring data for each of the clustered groups and their related critical lifetime prediction design parameters. A sensitivity analysis was conducted using the mathematical models, in order to produce graphical relations between the useful life and design parameters of a product. The validity of the suggested methodology was tested on electric motors and a gearbox as both these components have demonstrated great reuse potential. The results demonstrate that the methodology can assist designers in estimating the design life and associated design parameters with great accuracy, and subsequently aiding in a stratagem for reuse.

The main aim of this thesis is to develop a framework that allows an equipment manufacturer to forecast the demand of its aftermarket user-replaceable parts and prepare the challenges it may face from third party manufacturers. Some business practices ensuring the performance against third party part manufacturers are suggested in this thesis as a reference for a manufacturer to retain its customer and optimize its performance in aftermarket. The first part of this thesis focuses on the framework of a forecasting model for an equipment manufacturer which needs to provide aftermarket consumers to end users through the service or sales channels of the manufacturer. The forecasting model is carried out using the data from annual sales quantity of one model of equipments(known as install base) during the product life cycle. Each year during the product life cycle would have a different Rate of Replacement. The forecasting model aggregates the data from the install base and the rate of replacement to come up with an annual demand of the aftermarket replacement parts needed for the equipment. The second part of this thesis focuses on some practical business practices for the manufacture to ensure its performance in aftermarket. The designs of the products, logistics arrangement and selection of channels are also curial to its performance.

<p>This paper analyses the pricing on spare parts. Empirical studies have showed that manufacturers of durable goods make an unproportional large profit on its spare parts in relation to the revenue it generates. It is first showed that according to the standard economic model the price on spare part ought to be zero since the producer include an insurance in the price of the main good. Further it is showed that moral hazard alone do not explain the pricing found in the studies. Finally an analysis of whether consumers with present-biased preferences could be a possible explanation is made. The analysis finds that it is a possibility however somewhat unlikely.</p>

Cavity balancing is the process of altering the flow front within a cavity through thickness and design changes such that the desired fill pattern is achieved. The 2 dimensional (2D) cavity-balancing algorithm, developed by Lam and Seow [1] can only handle 2D geometry. This represents a major drawback as most, if not all of the practical injected parts are not 2D parts. To overcome this difficulty, the present investigation has developed a 2.5 dimensional (2.5D) cavity balancing optimization routine implemented within a 2.5 D finite elements domain. The aim of the automated cavity balancing routine is to reduce product development time and to improve product quality. This will lower the level of prerequisite expert knowledge necessary for successful mold and part design. The automated cavity balancing routine has been developed using the concept of flow paths. The hill-climbing algorithm of Lam and Seow is utilized but modified for the generation of flow paths for 2.5D parts. The algorithm has been implemented in a computer program running as an external loop to the MOLDFLOW software. Case studies are provided to demonstrate the efficiency of this routine. / Singapore-MIT Alliance (SMA)

This thesis presents a new high level robot programming system. The programming system can be used to construct strategies consisting of compliant motions, in which a moving robot slides along obstacles in its environment. The programming system is referred to as high level because the user is spared of many robot-level details, such as the specification of conditional tests, motion termination conditions, and compliance parameters. Instead, the user specifies task-level information, including a geometric model of the robot and its environment. The user may also have to specify some suggested motions. There are two main system components. The first component is an interactive teaching system which accepts motion commands from a user and attempts to build a compliant motion strategy using the specified motions as building blocks. The second component is an autonomous compliant motion planner, which is intended to spare the user from dealing with "simple" problems. The planner simplifies the representation of the environment by decomposing the configuration space of the robot into a finite state space, whose states are vertices, edges, faces, and combinations thereof. States are inked to each other by arcs, which represent reliable compliant motions. Using best first search, states are expanded until a strategy is found from the start state to a global state. This component represents one of the first implemented compliant motion planners. The programming system has been implemented on a Symbolics 3600 computer, and tested on several examples. One of the resulting compliant motion strategies was successfully executed on an IBM 7565 robot manipulator.