This thesis describes empirical research into project management of software development. Specifically, the aim of the research is to investigate how project managers monitor, control and coordinate software development tasks and how this is affected by changing environments, in this case increased organizational distance between the project manager and elements of the project team. Differing project environments allow investigation into which project management mechanisms are essential, which are required in specific circumstances and which may be useful but not necessarily essential. To explore how software development projects are monitored, controlled and coordinated, a broad range of literature from software development and other fields such as organization theory, supply chain management and automobile manufacture is examined to establish a consensus of the mechanisms of project monitoring, control and coordination and their classification into groups. To better understand how the different mechanisms may be selected in different circumstances, a range of contingencies is examined to deduce which of these contingencies may significantly affect the project management of software development projects. Outsourced and distributed software development projects are becoming more frequent than in the past with consequent effects on project management practices. Although there has been some research into the ways in which project managers monitor, control and coordinate software development projects, little of it has investigated how the mechanisms employed to do so may be affected by such factors as increasing organizational distance. If more were known about the ways in which changed project environments affected the selection and use of project management mechanisms, better responses to those environmental changes could be devised. This could also identify where tools could be developed to assist project management of outsourced and distributed projects. In this research, the term 'organizational distance' is used to describe the cultural, structural and administrative distance between the project manager and elements of the project team. Since there is limited information available on the concept of organizational distance, a new model is developed that encompasses the dimensions of distance that may be found in outsourced or globally distributed projects. A second model is then developed that relates changes in the factors of organizational distance with preferred choices of project management mechanism via project contingencies. Empirical data were collected by structured interviews with project managers who were currently engaged in software development within Sydney, Australia. The method of collecting the data provided both quantitative and qualitative data that enabled three separate ways to investigate the research questions. The empirical research found that project managers do not rely on a single mechanism to monitor, control or coordinate a software development project but employ multiple mechanisms. While the portfolio of mechanisms for both monitoring and control comprised a relatively narrow selection, the portfolio of coordination mechanisms was more diverse. Project monitoring mechanisms were employed to first detect any project problems then to respond to those problems. This contrasts with monitoring systems designed to provide all the information about both the existence and probable causes of project problems. Project control mechanisms reflected the origin of the control. The constraints imposed on the project by the organization and used by the project manager to direct the project tended to be outcome related, for example budget and schedule. The behaviour of the project team, even across significant organizational distances, was controlled through the use of project plans that determined when different tasks would be performed. Project coordination mechanisms reflected the different types of dependencies between software development activities. The most common was using a project work breakdown structure, expressed in the project schedule, to resolve sequential and pooled resource dependencies. Mutual dependencies tended to be resolved using interactive mechanisms such as co-location, conversations and meetings. The empirical evidence did not find any difference between co-located projects and distributed projects so far as the choice of project management mechanisms were concerned. Distributed and globally outsourced software development projects may encounter many difficulties that a fully co-located project does not, but the response to those difficulties appears to lie with the implementation of project management mechanisms and not their selection.

Thesis (Ph. D.)--Ohio State University, 2007. / Title from first page of PDF file. Includes bibliographical references (p. 243-249).

Thesis (PH.D.)--Michigan State University. Electrical and Computer Engineering, 2008. / Title from PDF t.p. (viewed on Aug. 11, 2009) Includes bibliographical references (p. 118-121). Also issued in print.

Ao se efetuar levantamentos utilizando a tecnologia GNSS, o processamento desses dados, bem como uma análise dos fatores oriundos dos processamentos e a correta interpretação dos resultados obtidos, consiste em fator primordial para se definir a qualidade de um levantamento. Contudo, os resultados estatísticos fornecidos pelos softwares comerciais após o processamento e ajustamento dos dados, apesar de garantir maior confiabilidade aos levantamentos, não fornecem a acurácia das coordenadas encontradas, apresentando apenas a sua precisão. Assim, este trabalho tem por objetivo final, fornecer uma tabela onde seja possível, através de uma comparação dos resultados oriundos dos processamentos, classificar a acurácia dos mesmos. Para tanto, foram efetuados os processamentos e ajustamentos de dados inerentes as quatro estações do ano, durante os anos de 2006, 2007 e 2008, e após adequada análise dos mesmos, foi implementada uma Tabela de Acurácia Recomendada, onde os profissionais da área de mensuração poderão, através de uma comparação entre os relatórios de processamento fornecidos e os itens da referida tabela, efetuar a classificação da acurácia dos trabalhos de interesse. / When making surveys using GNSS technology, the processing of these data, as well as an analysis of the factors resulting from the processing and the correct interpretation of results, is to prime factor for defining the quality of a survey. However, the statistics provided by commercial software after processing and adjustment of the data, while ensuring high reliability of surveys, do not provide the accuracy of the coordinates found, showing only its precision. This paper aims to provide a table where possible, by comparing the results from the processing, The classification accuracy of them. To do so, we made adjustments and the processing of data relating the four seasons, during the years 2006, 2007 and 2008, and after proper analysis of the same purpose, a Table of Accuracy Recommended where professionals in the area of measurement may, through a comparison between the reports provided and processing the items of the tariff, to perform the classification accuracy of the work of interest.

Orientador: Nilson Tadeu Mascia / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Civil, Arquitetura e Urbanismo / Made available in DSpace on 2018-08-19T08:46:18Z (GMT). No. of bitstreams: 1 Moraes_ReinaldoWashington_M.pdf: 3150718 bytes, checksum: a564e4cc9fe162e6f88384d199323bc9 (MD5) Previous issue date: 2011 / Resumo: Devido à crescente utilização da madeira no Brasil, estudos tecnológicos são aprimorados nos últimos anos sendo verificada a necessidade de pesquisas direcionadas a madeiras de reflorestamento e das técnicas construtivas por meio de diversas composições com outros materiais de construção. As pesquisas nessa linha contribuem para a diminuição da pressão de exploração de novas áreas proporcionando novos usos para a madeira de reflorestamento como material estrutural pela indústria da construção civil. Estudos a respeito das propriedades mecânicas e físicas de madeiras e suas composições importantes para o aprimoramento e desenvolvimento dessas novas tecnologias. No presente trabalho, o estudo de vigas de madeira laminada colada reforçadas com fibras de sisal, confeccionadas de madeiras de reflorestamento, é realizado com ênfase no contexto do módulo de elasticidade. São apresentados compostos laminados e alguns tipos de reforços utilizados comumente na engenharia de estruturas, voltando à pesquisa, principalmente para as fibras naturais. Além disso, são descritos alguns conceitos de elasticidade, leis constitutivas e as transformações de coordenadas sob a ótica de importantes autores. É desenvolvida uma formulação cuja análise é feita utilizando-se um modelo de transformação tridimensional, não convencional na literatura pesquisada, sendo possível então a aplicação da teoria desenvolvida no modelo de vigas laminada colada proposto no trabalho. Os resultados obtidos mostram que as transformações tridimensionais são de grande valia ao estudo da anisotropia dos materiais e, em particular, para o material madeira e suas composições estruturais / Abstract: Due to the increasing use of wood in Brazil, technological studies are being improved in recent years seen the need for research aimed at reforestation and timber construction techniques through a variety of compositions with other building materials. Research in this line helps reduce the pressure to exploit new areas providing new uses for wood as a structural material for reforestation by the construction industry. Studies of mechanical and physical properties of wood composites and their importance to the improvement and development of these new technologies. In this study, the study of glued laminated wood beams reinforced with sisal fibers, made of wood from reforestation is carried out with emphasis in the context of the modulus of elasticity. Composite laminates are presented and some types of reinforcements commonly used in structural engineering, returning to research, especially for natural fibers. In addition, we describe some concepts of elasticity, constitutive laws and the coordinate transformations from the perspective of important authors. It developed a formulation whose analysis is done using a transformation model three-dimensional, non-conventional literature, it is possible then the application of the theory developed in the model proposed in glued laminated beams work. The results show that the three-dimensional transformations are of great value to study the anisotropy of materials and, in particular, for the wood material and its structural composition / Mestrado / Estruturas / Mestre em Engenharia Civil

Abstract This thesis describes position-sensitive devices (PSDs) and optical sensor systems suitable for industrial tracking and displacement sensing applications. The main application areas of the proposed sensors include automatic pointing of a rangefinder beam and measuring the lateral displacement of an object. A conventional tracking sensor is composed of a laser illuminator, a misfocused quadrant detector (QD) receiver and a corner cube retroreflector (CCR) attached to the target. The angular displacement of a target from the receiver optical axis is detected by illuminating the target and determining the direction of the reflection using the QD receiver. The main contribution of the thesis is related to the modifications proposed for this conventional construction in order to make its performance sufficient for industrial applications that require a few millimetre to submillimetre accuracy. The work includes sensor optical construction modifications and the designing of new types of PSDs. The conventional QD-based sensor, although electrically very sensitive, is not considered optimal for industrial applications since its precision is severely hampered by atmospheric turbulence due to the misfocusing needed for its operation. Replacing the CCR with a sheet reflector is found to improve the precision of the conventional sensor construction in outdoor beam pointing applications, and is estimated to allow subcentimetre precision over distances of up to 100 m under most operating conditions. Submillimetre accuracy is achievable in close-range beam pointing applications using a small piece of sheet reflector, coaxial illumination and a focused QD receiver. Polarisation filtering is found to be effective in eliminating the main error contributor in close-range applications, which is low reflector background contrast, especially in cases when a sheet reflector has a specularly reflecting background. The tracking sensor construction is also proposed for measuring the aiming trajectory of a firearm in an outdoor environment. This time an order of magnitude improvement in precision is achieved by replacing the QD with a focused lateral effect photodiode (LEP). Use of this construction in cases of intermediate atmospheric turbulence allows a precision better than 1 cm to be achieved up to a distance of 300 m. A method based on averaging the positions of multiple reflectors is also proposed in order to improve the precision in turbulence-limited cases. Finally, various types of custom-designed PSDs utilising a photodetector array structure are presented for long-range displacement sensing applications. The goal was to be able to replace the noisy LEP with a low-noise PSD without compromising the low turbulence sensitivity achievable with the LEP. An order of magnitude improvement in incremental sensitivity is achievable with the proposed array PSDs.

The work deals with the issues of network coordinate systems. The first chapter is devoted to a brief analysis of this systems and requirements placed on them. Furthermore, it deals with features of Vivaldi algorithm and its three versions: a centralized algorithm, algorithm with constant and adaptive time step. Subsequently it describes the GNP system and eash step of Nelder-Mead method for nodes localization. Next to it a developed simulation library is presented with its own graphical user interface for testing of these algorithms, which is capable of handling RTT responses database amongst individual IP network members. Simulations performed on data from PlanetLab network are evaluated in the conclusion.

This thesis has two aims. The first one is the investigation and analysis of the errors occurring in the measurements with pulsed time-of-flight (TOF) terrestrial laser scanners (TLS). A good understanding of the error sources and the relationships between them is necessary to secure the data accuracy. We subdivide these errors into four groups: instrumental, object-related, environmental and methodological. Based on our studies and the results obtained by other researchers, we have compiled an error model for TLS, which is used to estimate the single-point coordinate accuracy of a point in the point cloud, transformed to the specified coordinate system. The second aim is to investigate systematic instrumental errors and performance of three pulsed TOF laser scanners – Callidus 1.1, Leica HDS 3000 and Leica HDS 2500 – and to develop calibration procedures that can be applied by the users to determine and correct the systematic errors in these instruments. The investigations have been performed at the indoor 3D calibration field established at KTH and outdoors. The systematic instrumental errors, or calibration parameters, have been estimated in a self-calibration according to the parametric least-squares adjustment in MATLAB®. The initial assumption was that the scanner instrumental errors are similar to those in a total station. The results have shown that the total station error model is applicable for TLS as a first approximation, but additional errors, specific to the scanner design, may appear. For example, we revealed a significant vertical scale error in the scanner Callidus 1.1, caused by the faults of the angular position sensor. The coordinate precision and accuracy of the scanners, estimated during the self-calibration, is at the level of several millimetres for Callidus 1.1 and Leica HDS 3000, and at the submillimetre level for Leica HDS 2500. In other investigations, we revealed a range drift of up to 3 mm during the first few hours of scanning, presumably due to the changes in the temperature inside the scanners. The angular precision depends on the scanner design (“panoramic” or “camera-like”), and the angular accuracy depends on the significant calibration parameters in the scanner. Investigations of the influence of surface reflectance on the range measurements have shown that the indoor illumination and surface wetness have no tangible influence on the results. The type of the material does not affect, in general, the ranging precision for Callidus 1.1, but it affects the ranging precision and accuracy of the scanners Leica HDS 3000 and Leica HDS 2500. The reason may be different wavelength and, possibly, different design of the electronics in the laser rangefinders. Materials with high reflectance and those painted with bright “warning” colours may introduce significant offsets into the measured ranges (5 – 15 cm), when scanned from close ranges at normal incidence with the scanner Leica HDS 3000. “Mixed pixels” at the object edge may introduce a range error of several centimetres, on the average, depending on the type of the material. This phenomenon leads also to the distortions of the object size, which may be reduced by the removal of the “mixed pixels” based on their intensity. The laser beam intensity recorded by the scanner tends to decrease with an increased incidence angle, although not as assumed by the popular Lambertian reflectance model. Investigations of the scanner Leica HDS 2500 outdoors have revealed no significant influence of the “normal” atmospheric conditions on the range measurements at the ranges of up to 50 m. Finally, we have developed and tested two simple procedures for the calibration of the vertical scale (and vertical index) error and zero error in laser scanners. We have also proposed an approach for the evaluation of the coordinate precision and accuracy in TLS based on the experiences from airborne laser scanning (ALS). / QC 20101123

Many key problems in machine learning and data science are routinely modeled as optimization problems and solved via optimization algorithms. With the increase of the volume of data and the size and complexity of the statistical models used to formulate these often ill-conditioned optimization tasks, there is a need for new efficient algorithms able to cope with these challenges. In this thesis, we deal with each of these sources of difficulty in a different way. To efficiently address the big data issue, we develop new methods which in each iteration examine a small random subset of the training data only. To handle the big model issue, we develop methods which in each iteration update a random subset of the model parameters only. Finally, to deal with ill-conditioned problems, we devise methods that incorporate either higher-order information or Nesterov’s acceleration/momentum. In all cases, randomness is viewed as a powerful algorithmic tool that we tune, both in theory and in experiments, to achieve the best results. Our algorithms have their primary application in training supervised machine learning models via regularized empirical risk minimization, which is the dominant paradigm for training such models. However, due to their generality, our methods can be applied in many other fields, including but not limited to data science, engineering, scientific computing, and statistics.

No description available.