Includes abstract. / Includes bibliographical references (p. 89-90). / This thesis proposes the use of data mining tools within an operations research process, allowing the integration of ever increasing amounts of data collected worldwide. It is further argued that the wealth of information delivered by DM tools, with their strong visual emphasis, can be used by enhance the transfer of knowledge of stakeholders. The discipline of operations research could benefit greatly from the methods offered within the field of data mining, used to analyse the ever increasing amounts of data collected worldwide. However, the subject also offers a wealth of information that could aid in decision making, along with visual representations of this information that might assist in the transferral of knowledge to problem stakeholders. The advantages offered by data mining are not limited to problem contexts containing high-quality data, but could also assist within the development contexts containing high-quality data, but could also assist within the development context where traditionally resources and relevant skills are scarce. The benefits of data mining within this context are illustrated through the use of municipal performance data supplied by the Department of Local Government and Housing in the Western Cape of South Africa. The results of these analyses are presented to the department in order to assess the potential contribution of data mining to decisions surrounding municipal support.

Includes bibliographical references (leaves 132-137). / Malaria ranks among the world's most important tropical parasitic diseases with world prevalence figures between 350 and 550 million clinical cases per annum. [WHO, 2008a] 'Treatment and prevention of malaria places a considerable burden on struggling economies where the disease is rampant. Research in malaria does not stop as the change in response to antimalarial drug treatment requires the development of new drugs and innovation in the use of old drugs. This thesis focused on building a model of the spread of resistance to Sulfadoxine/Pyrimethamine (SP) in a setting where both SP and SP in artemisinin-based combination therapy (ACT) are the first line therapies for malaria. The model itself is suitable to any low transmission setting where antimalarial drug resistance exists but the country of choice in this modeling exercise was Mozambique. The model was calibrated using parameters specific to the malaria situation in Mozambique. This model was intended to be used to aid decision making in countries where antimalarial drug resistance exists to help prevent resistance spreading to such an extent that drugs lose their usefulness in curing malaria. The modeling technique of choice was differential equation modeling; a simulation technique that falls under the System Dynamics banner in the Operations Research armamentarium. It is a technique that allowed the modeling of stocks and flows that represent different stages or groupings in the disease process and the rate of movement between these stages respectively. The base model that was built allowed infected individuals to become infectious, to be treated with SP or ACT and to be sensitive to or fail treatment. Individuals were allowed a period of temporary immunity where they would not be reinfected until the residual SP had been eliminated from their bloodstream. The base model was then further developed to include the pharmacokinetic properties of SP where individuals were allowed to be reinfected with certain strains of infection given the level of residual drug in their bloodstream after their current infection had been cleared. The models used in this thesis were built with idea of expanding on previous models and using available data to improve parameter estimates. The model at its core is similar to the resistance model used in Koella and Antia [2003] where differential equation modeling was used to monitor a population as it became infected with a sensitive or resistant infection and then University of Cape Town recovered. The inclusion in the model of the PK component was derived from Prudhomme-O'Meara et al. [2006] where individuals could be reinfected depending on the residual drug in their bloodstream. Rather than modeling simply sensitive and resistant infections, mutations categories were used as was the case in Watkins et al. [2005] population genetics model. The use of mutation categories allowed one to use parameters specific to these categories rather than the sensitive/resistant stratification and this is particularly relevant in Mozambique where all mutation categories still exhibit some degree of sensitivity to treatment i.e. total resistance has not yet developed for any particular mutation category. The last adaptation of the model was to use gametocyte information directly to determine human infectiousness rather than through using a gametocyte switching rate (constant multiplier used to convert parasite density to gametocyte density) as was done in Pongtavompinyo [2006]. The models developed in this thesis found that the existing vector control and drug policy in Mozambique had the major effect of decreasing total prevalence of malaria by approximately 70% in the 11 year period. The distribution of Res3 (presence of DHFR triple) and Res5 (presence of DHFR triple and DHPS double) infections changed over the 11 year period with Res3 infections initially increasing and then decreasing while Res5 infections started low and increased to overtake Res3 infections. The timing of the change in this composition of infection corresponds with the introduction of ACT and thus it appears that the use of ACT prompted the increased prevalence of quintuple parasites over DHFR triple and sensitive parasites. The total number of failures decreased substantially after the introduction of ACT to 17% of its previous level. The results of the base model corresponded with the observed data from the SEACAT study in terms of the magnitude and the trends of the impact of the change to ACT policy, but underestimated the impact of the vector control strategies compared to rapid effect noted in Sharp et al. [2007]. The Scenario testing of the base model showed that vector control is an effective strategy to reduce prevalence and that it is sensitive to the time at which the control is started as it decreased prevalence very gradually. The Scenario testing of the base model also showed that the introduction of ACT in Mozambique had a greater impact on reducing prevalence and that the start time of the ACT strategy did not decrease the effect on prevalence though earlier start times decreased the total number of resistance cases. The ratio of Res5 to Res3 infections increased faster when ACT was the treatment policy than when SP was the policy. Thus higher values of this ratio are associated with ACT being the treatment strategy in place. Thus differential equation modeling is an effective modeling tool to capture the spread of disease and to test the effects of policy interventions as it allows one to assess these effects on populations and averages out individual-level intricacies to better inform policy decisions.

Goodman derived the complex Hishart distribution with the aid of characteristic functions and Fourier transforms. From this the distribution of the complex rectangular co-ordinates were derived as an application of the complex Hishart distribution. In the present paper we give a direct and simplified method of deriving the distribution of the complex rectangular co-ordinates. From this distribution the complex Hishart distribution will be derived as an application. Some properties of the distribution of the complex rectangular coordinates will be given. In addition some applications and in particular the application of the distribution in the derivation of the distribution of the "complex" generalized variance will also be given.

This thesis attempts to isolate the major markets for investment outside South Africa, and to determine what proportion of a South African investor's capital, if any, should be held in foreign securities under different possible restrictions laid down by the South African Reserve Bank. The main argument advanced in favour of foreign investment is risk reduction through diversification.

abstract: Conservation planning is fundamental to guarantee the survival of endangered species and to preserve the ecological values of some ecosystems. Planning land acquisitions increasingly requires a landscape approach to mitigate the negative impacts of spatial threats such as urbanization, agricultural development, and climate change. In this context, landscape connectivity and compactness are vital characteristics for the effective functionality of conservation reserves. Connectivity allows species to travel across landscapes, facilitating the flow of genes across populations from different protected areas. Compactness measures the spatial dispersion of protected sites, which can be used to mitigate risk factors associated with species leaving and re-entering the reserve. This research proposes an optimization model to identify areas to protect while enforcing connectivity and compactness. In the suggested projected area, this research builds upon existing methods and develops an alternative metric of compactness that penalizes the selection of patches of land with few protected neighbors. The new metric is referred as leaf because it intends to minimize the number of selected areas with 1 neighboring protected area. The model includes budget and minimum selected area constraints to reflect realistic financial and ecological requirements. Using a lexicographic approach, the model can improve the compactness of conservation reserves obtained by other methods. The use of the model is illustrated by solving instances of up to 1100 patches. / Dissertation/Thesis / Masters Thesis Industrial Engineering 2019

Thesis (Ph. D.)--Massachusetts Institute of Technology, Sloan School of Management, Operations Research Center, 2010. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 201-213). / In this thesis, we examine a recent paradigm for solving dynamic optimization problems under uncertainty, whereby one considers decisions that depend directly on the sequence of observed disturbances. The resulting policies, called recourse decision rules, originated in Stochastic Programming, and have been widely adopted in recent works in Robust Control and Robust Optimization; the specific subclass of affine policies has been found to be tractable and to deliver excellent empirical performance in several relevant models and applications. In the first chapter of the thesis, using ideas from polyhedral geometry, we prove that disturbance-affine policies are optimal in the context of a one-dimensional, constrained dynamical system. Our approach leads to policies that can be computed by solving a single linear program, and which bear an interesting decomposition property, which we explore in connection with a classical inventory management problem. The result also underscores a fundamental distinction between robust and stochastic models for dynamic optimization, with the former resulting in qualitatively simpler problems than the latter. In the second chapter, we introduce a hierarchy of polynomial policies that are also directly parameterized in the observed uncertainties, and that can be efficiently computed using semidefinite optimization methods. The hierarchy is asymptotically optimal and guaranteed to improve over affine policies for a large class of relevant problems. To test our framework, we consider two problem instances arising in inventory management, for which we find that quadratic policies considerably improve over affine ones, while cubic policies essentially close the optimality gap. In the final chapter, we examine the problem of dynamically pricing inventories in multiple items, in order to maximize revenues. For a linear demand function, we propose a distributionally robust uncertainty model, argue how it can be constructed from limited historical data, and show how pricing policies depending on the observed model mis-specifications can be computed by solving second-order conic or semidefinite optimization problems. We calibrate and test our model using both synthetic data, as well as real data from a large US retailer. Extensive Monte-Carlo simulations show 3 that adaptive robust policies considerably improve over open-loop formulations, and are competitive with popular heuristics in the literature. / by Dan Andrei Iancu. / Ph.D.

No description available.

No description available.

No description available.

No description available.