We adopt a Bayes linear approach to tackle design problems with many variables cross-classified in many ways. We investigate designs where we wish to sample individuals belonging to different groups, exploiting the powerful properties of the adjustment of infinitely second-order exchangeable vectors. The types of information we gain by sampling are identified with the orthogonal canonical directions. We show how we may express these directions in terms of the different factors of the model. This allows us to solve a series of lower dimensional problems, through which we may identify the different aspects of our adjusted beliefs with the different aspects of the choice of design, leading both to qualitative insights and quantitative guidance for the optimal choice of design. These subproblems have an interpretable form in terms of adjustment upon subspaces of the full problem and remain valid when we consider adjusting the underlying population structure and also for predicting future observables from past observation. We then examine the adjustment of finitely second-order exchangeable vectors, and show that the adjustment shares the same powerful properties as the adjustment in the infinite case. We show how if the finite sequence of vectors is extendible, then the differences in the adjustment of the sequence is quantitatively the same for all sequence lengths and it is easy to compare the qualitative differences. Extending to an infinite sequence allows us to draw comparisons between the finite and infinite modelling. Such comparisons may also be made when we consider sampling individuals belonging to different groups, where each group contains only a finite number of individuals.

Simularam-se 1000 ensaios para cada uma das 112 combinações de 4 populações (tratamentos), englobando casos de populações iguais e casos com diferenças em m, k ou ambos. Deve-se ter cuidado ao aplicar transformações de dados, particularmente se não há homogeneidade de k. A estatística C(α) proposta por BARNWAL & PAUL (1988), mostrou alguma robustez para valores não homogêneos de k, conduzindo a resultados equivalentes àqueles obtidos usado dados não transformados. A análise de variância, usado o teste de mínimo qui-quadrado XU2 mostrou ser viesado, superestimando valores quando a matriz de variâncias e covariâncias é desconhecida. Se a matriz de variâncias e covariâncias é conhecida, os resultados são equivalentes a aqueles obtidos dos dados originais. Resultados similares foram obtidos para populações menores, n=10, quando um poder decrescente do teste foi detectado. Foram escolhidos 20 casos e simularam-se 1000 ensaios para cada caso. / The well-known negative binomial distribution is quite frequently used to interpret counting variables, through different techniques. In order to compare these techniques, four populations of size n=50 were computer-generated for different values of the parameters m and k, using NORMAN & CANNON (1972) procedure. Comparisons of the transformations of variables, as suggested by BARBOSA (1985), were used. For that, 1000 essays were simulated for each one of the 112 combinations of 4 populations. This covered equal and different populations with respect to the parameters m, k or both. In conclusion, for some values of the parameters m and k, there is no necessity of any data transformation, particularly if depending of k. Statistics like C(α) proposed by BARNWAL & PAUL (1988) showed some robustness for non-homogeneous values of k, leading to equivalent results to that ones obtainned using untransformed data. The analysis of variance, using the minimum chi-square test U2 showed to be biased superestimating values when the variance-covariance matrix is unknown. If the variance-covariance matrix is knew the results are equivalent from those obtainned from original data. Similar results were obtainned for smaller populations, n=10, when a decreasing power of the tests was detected. In such a case 20 combinations and 1000 simulations for each combination were performed

No description available.

A Dissertation Submitted of the Faculty of Science, University of the Witwatersrand, Johannesburg in Partial Fulfilment of the Requirements for the Degree of Master of Science / Let 4 be the class of pairs of loc ..My onvex spaces (X,V) “h ‘ch are such that every closed graph linear ,pp, 1 from X into V is continuous. It B is any class of locally . ivex l.ausdortf spaces. let & w . (X . (X.Y) e 4 for ,11 Y E B). " ‘his expository dissertation, * (B) is investigated, firstly i r arbitrary B . secondly when B is the class of C,-complete paces and thirdly whon B is a class of locally convex webbed s- .ces

Este trabalho visa obter um plano de corte economicamente ótimo para florestas de eucaliptos. A técnica utilizada é a programação linear. O objetivo do modelo desenvolvido é a maximização do retorno sobre os investimentos. Para tanto, foram calculados o Valor Líquido Presente (VLP) de todas as alternativas de manejo dadas ao modelo, neste cálculo foi considerado como parte integrante do fluxo de caixa o Valor Esperado da Terra (VET). Na determinação do plano de manejo mais adequado, foram analisados quatro diferentes cenários de planejamento dos pontos de vista econômico e operacional. Os quatro cenários foram: 1) oscilação livre da produção anual; 2) suprimento anual de madeira para celulose superior ao limite de 40.000 m3/ano; 3) suprimento anual de madeira para celulose inferior ao limite máximo de 100.000 m3/ano; 4) suprimento anual de madeira para celulose restrito a intervalos. Os resultados demonstraram que, independente da estratégia de manejo obtida, dentro das condições etipuladas, a produção florestal não é deficitária. A estratégia de manejo de corte considerada mais apropriada foi a do cenário cujo fluxo de produção ficou restrito a intervalos / The purpose of this paper was to obtain, an optimal economic planning for management and logging of Eucalypt forests. The technique utilized is linear programming. The objective of the modelo was to maximize the return over the investiments. Therefor the Net Present Value (NPV) of all management alternative conditions given to were calculated the model. In this computation the Bare Land Value (BLV) was considered part of the cash flow. For determination of the most suitable management planning to the mill supply, four different scenarios of economical and operational planning were analised. The scenarios were: 1) free oscillation to the anual productions of wood; 2) Annuak wood supply for pulpwood above minimum limit (40.000 m3/year); 3) annual wood supply for pulpwood below the maximum limit the 100.000 m3/year; 4) annual woos supply for pulpwood restricted to pre-fixed intervals. The results showed that, independent of management strategy obtained under the considered adjusted conditions, the forest production was not deficient. The best management alternative, considering the technological and operational limitations, was that from the scenary in which the production flow was restricted to pre-fixed intervals

by Ma Mang Fai. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1989. / Bibliography: leaf 77.

Thesis (M.Phil.)--Chinese University of Hong Kong. / Bibliography: leaf [47]

Linear induction machines currently play a relatively minor role in the industrial world. This is partly due to relatively high production costs, complexity of construction and the Inability to apply standard mass production techniques. The aim of this thesis is to investigate the design of linear machines that are cheaper and faster to produce, and that may easily be mass-produced This thesis principally concerns the use of concentrated winding linear stators. These are cheap and easy to manufacture and can be easily mass-produced. However, high levels of negative harmonics make them unsuitable for use with simple sheet rotors.

Digitized by Kansas Correctional Industries

Made available in DSpace on 2019-03-29T23:27:59Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-09-27 / Transport of goods has represented an important role in economic development throughout the history and ship containerization brought great advantages. Its invention in mid-1950s brought down the cost of transport and reduced time for loading and unloading cargo. Consequently, it increased efficiency of port working and reduced handling cargo to hours instead of weeks, as before. However, the good use of containerization involves new and specialized logistic process, a number of technologies and automated systems to handle a great number of containers and even greater volume of cargo. To answer these requirements, computation appears as important tool. The described scenary has been treated in academic literature as the Container Loading Problem (CLP), with some variants. It is necessary consider practical requirements, for example the stability of cargo or weight distribution. The last one is of vital importance since the position of the centre of gravity of cargo affects the stability during its transport. When desconsidered, it could result in damage to cargo or vehicle. During our research, we were specially interested in this requirement. But, in order compare the found solutions with other ones, we proposed a methodology to measures the weight distribution. So, to the described problem, specifically the Knapsack Loading Problem (3D-KLP), this work presents a methodology that not only maximizes the packed cargo volume but also optimizes the weight distribution, its great contribution. Mainly if we consider that the cargo to be packed is composed by items with different densities, which turns the problem more difficult. The present methodology is composed by two phases with distinct goals. The first phase is concerned with maximize the weight distribution combining a search algorithm, the backtracking, with heuristics that solve integer linear programming models. The second phase executes a Genetic Algorithm to maximize the weight distribution of previously packed cargo. We also present a justification for why genetic algorithm was used in our methodology. An additional application was made to solve cutting problems. This class of problems occurs in various industrial process, when it is necessary to cut different types of material as glass, wood or parper, with a minimum of waste. We use a well-known benchmark test to compare our results with other approaches. This work also presents a case study of our implementation using some real data in a factory of stoves and refrigerators in Brazil. It shown promising results in reduced time. Keywords: Container Loading Problem, Knapsack Loading Problem, Weight Distribution, Integer Programming, Backtracking, Genetic Algorithms. / O transporte de carga tem representado um papel fundamental no desenvolvimento econômico no decorrer da história e a conteinerização trouxe grandes vantagens. Seu advento reduziu os custos de transporte bem como o tempo de carga. Portanto, aumentou a eficiência do trabalho em portos e reduziu o tempo necessário para operações com carga para horas, ao invés de semanas como anteriormente. Contudo, o bom uso dos contêineres involve novos e especializados processos logísticos, uma grande quantidade de tecnologias além de sistemas automatizados para manipular uma elevada quantidade de contêineres e ainda maior volume de carga. Para atender a estes requisitos, computação aparece como uma importante ferramenta. O cenário descrito tem sido tratado na literatura acadêmica como o Problema de Carregamento de Contêiner (CLP, do inglês Container Loading Problem), com algumas variantes. é também necessário considerar requisitos práticos como, por exemplo, a estabilidade da carga ou distribuição do peso. Este último de vital importância uma vez que o centro de gravidade da carga afeta a estabilidade durante seu transporte. Se descosiderado, pode-se danificar tanto a carga como o veículo. Durante nossa pesquisa, nós estivemos especialmente interessados neste requisito. E a fim de comparar a qualidade dos resultados obtidos, propusemos uma maneira de mensurar a distribuição do peso. Portanto, dado o problema descrito, especificamente o 3D Knapsack Loading Problem, este trabalho apresenta um algoritmo que não apenas maximiza o volume total carregado mas também otimiza a distribuição do peso da carga, sua grande contribuição. Principalmente se considerarmos que a carga é composta de itens com diferentes valores de densidade, o que torna o problema ainda mais difícil. A metodologia consiste em duas fases com objetivos diferentes. A primeira fase ocupa-se em maximizar o volume carregado por combinar um algoritmo de busca, o backtracking, com heurísticas que resolvem modelos de programação linear inteira. A segunda fase executa um algoritmo genético para maximizar a distribuição do peso da carga previamente colocada. Apresentamos também uma justificativa do porque algoritmo genéticos foram usados em nossa metodologia. Uma aplicação adicional foi feita para resolver problemas de corte. Esta classe de problemas ocorre em vários processos industriais, quando é necessário cortar diferentes tipos de materiais, como vidro, madeira ou papel, com um mínimo de desperdício. A fim de comparação, usamos bibliotecas de teste bem conhecidas na literatura e um estudo de caso usando informações reais de uma fábrica de fogões e geladeiras no Brasil. São apresentados resultados promissores alcançados em tempo reduzido. Palavras-chave: Problema de Carregamento de Contêiner, Knapsack Loading Problem, Distribuição do Peso, Programação Linear Inteira, Backtracking, Algoritmos Genéticos.