Ett undersöknings- och utvecklingsarbete för att lösa ett problem med att dränkbara pumpar i ett vattenavrinningssystem gick sönder med perioder på två år i genomsnitt, utfördes i samarbete med Cementa AB i Skövde. Orsak till pumpars haveri söktes och fanns vara bristande rutiner och kunskap om det underhåll pumparna krävde. För att lösa detta problem utvecklades riktlinjer för nyinköp av torruppställda pumpar för att möjliggöra kontinuerligt underhåll. Då möjligheter för placering av torruppställd pump saknades utvecklades ett pumpfundament för placering av torruppställd pump. Krav för utvecklingsarbetet togs fram i samarbete med Cementas underhållsavdelning och teoretisk dimensionering av dränkbara länspumpars dåvarande volymflödeskapacitet utfördes. Krav utvärderades och viktades med hjälp av parvis jämförelse. Dimensionering och kontroll av hållfasthet för utvecklat pumpfundamentet utfördes med hjälp av Finita Element Analyser i programvaran Pro/Engineer Creo 1.0 Mechanica. Kontroll av hållfasthet i infästning av pumpfundamentet samt svetsfogar utfördes analytiskt. Arbetet resulterade i en rekommendation till Cementa AB i Skövde att ta in offerter på nya torruppställda pumpar med hjälp av utvecklade riktlinjer och att tillverka det pumpfundament som tagits fram inom ramen för examensarbetet, för att placera nya pumpar på. Att noggrant följa de underhållsinstruktioner som pumpar har och att underlätta för personal att utföra detta underhåll ansågs kunna bidra till att pumpar skulle få en längre och mer ekonomisk livslängd. / A development project to solve problems with why submersible pumps in a run-off system broke down with periods of two years, on average, was performed in collaboration with Cementa AB in Skövde. Reason for the pumps breakdowns was searched and found to be inadequate procedures and missing knowledge of the maintenance required on the pumps. To solve this problem, guidelines for the purchase of new dry pit pumps were developed to allow for continuous maintenance. As the possibilities of placing a dry well pump did not exist at Cementa, a pump foundation was developed. Requirements for the development work were produced in cooperation with Cementas maintenance department and theoretical dimensioning of the submersible bilge pumps volume flow capacity was performed. Requirements were evaluated and weighted using Pairwise comparison. The design and control of the strength of the developed pump foundation was performed using finite element analysis in the software Pro/Engineer Creo 1.0 Mechanica. Controls of the strength of the attachment of the pump foundation and welds were performed analytically. The work resulted in a recommendation to Cementa AB in Skövde to bring in quotes on the new dry-pit pumps using the developed guidelines and to manufacture the pump foundation developed within the framework of the thesis. Cementa was also recommended to carefully follow the maintenance instructions for pumps and make it easier for staff to perform this maintenance. This was recommended to ensure that new pumps would have a longer and more economical lifetime.

Ranking a set of candidates or items from pair-wise comparisons is a fundamental problem that arises in many settings such as elections, recommendation systems, sports team rankings, document rankings and so on. Indeed it is well known in the psychology literature that when a large number of items are to be ranked, it is easier for humans to give pair-wise comparisons as opposed to complete rankings. The problem of ranking from pair-wise comparisons has been studied in multiple communities such as machine learning, operations research, linear algebra, statistics etc., and several algorithms (both classic and recent) have been proposed. However, it is not well under-stood under what conditions these different algorithms perform well. In this thesis, we aim to fill this fundamental gap, by elucidating precise conditions under which different algorithms perform well, as well as giving new algorithms that provably perform well under broader conditions. In particular, we consider a natural statistical model wherein for every pair of items (i; j), there is a probability Pij such that each time items i and j are compared, item j beats item i with probability Pij . Such models, which we summarize through a matrix containing all these pair-wise probabilities, have been used explicitly or implicitly in much previous work in the area; we refer to the resulting matrix as the pair-wise preference matrix, and elucidate clearly the crucial role it plays in determining the performance of various algorithms. In the first part of the thesis, we consider a natural generative model where all pairs of items can be sampled and where the underlying preferences are assumed to be acyclic. Under this setting, we elucidate the conditions on the pair-wise preference matrix under which popular algorithms such as matrix Borda, spectral ranking, least squares and maximum likelihood under a Bradley-Terry-Luce (BTL) model produce optimal rankings that minimize the pair-wise disagreement error. Specifically, we derive explicit sample complexity bounds for each of these algorithms to output an optimal ranking under interesting subclasses of the class of all acyclic pair-wise preference matrices. We show that none of these popular algorithms is guaranteed to produce optimal rankings for all acyclic preference matrices. We then pro-pose a novel support vector machine based rank aggregation algorithm that provably does so. In the second part of the thesis, we consider the setting where preferences may contain cycles. Here, finding a ranking that minimizes the pairwise disagreement error is in general NP-hard. However, even in the presence of cycles, one may wish to rank 'good' items ahead of the rest. We develop a framework for this setting using notions of winners based on tournament solution concepts from social choice theory. We first show that none of the existing algorithms are guaranteed to rank winners ahead of the rest for popular tournament solution based winners such as top cycle, Copeland set, Markov set etc. We propose three algorithms - matrix Copeland, unweighted Markov and parametric Markov - which provably rank winners at the top for these popular tournament solutions. In addition to ranking winners at the top, we show that the rankings output by the matrix Copeland and the parametric Markov algorithms also minimize the pair-wise disagreement error for certain classes of acyclic preference matrices. Finally, in the third part of the thesis, we consider the setting where the number of items to be ranked is large and it is impractical to obtain comparisons among all pairs. Here, one samples a small set of pairs uniformly at random and compares each pair a fixed number of times; in particular, the goal is to come up with good algorithms that sample comparisons among only O(nlog(n)) item pairs (where n is the number of items). Unlike existing results for such settings, where one either assumes a noisy permutation model (under which there is a true underlying ranking and the outcome of every comparison differs from the true ranking with some fixed probability) or assumes a BTL or Thurstone model, we develop a general algorithmic framework based on ideas from matrix completion, termed low-rank pair-wise ranking, which provably produces an good ranking by comparing only O(nlog(n)) pairs, O(log(n)) times each, not only for popular classes of models such as BTL and Thurstone, but also for much more general classes of models wherein a suitable transform of the pair-wise probabilities leads to a low-rank matrix; this subsumes the guarantees of many previous algorithms in this setting. Overall, our results help to understand at a fundamental level the statistical properties of various algorithms for the problem of ranking from pair-wise comparisons, and under various natural settings, lead to novel algorithms with improved statistical guarantees compared to existing algorithms for this problem.

This masters thesis is focused around the comparison of time dependent video grayscale conversion methods based on a user experiment. The test methodology is based on the pairwise comparison method 2AFC. It is composed of two test variants: a test with a reference video and a test without a reference video. The coefficients of agreement, consistency and correlation are utilized in the result analysis. The testing was done on 60 subjects which have done 7200 pairwise comparisons. The test results show that the time dependent method Hu14 is the most universal. Time dependent method Kim09 was bested by some non-dependent methods. The results also indicate that there is some correlation between the two test variants and that the choice of the input video can have impact on the method performance. The main contribution of this thesis is that non-dependent methods can under certain circumstances rival the performance of dependent methods and that the addition of a reference video did not have a meaningful impact on the test subjects judgements.

The current decision-making processes involved in Environmental Impact Assessments (EIAs) in South Africa suffer from a lack of coherence and do not include evaluation of trade-offs between qualitative and quantitative impacts, as well as environmental, economic, and social dimensions. In addition, insufficient capacity and knowledge among authorities, a lack of objectivity among Environmental Assessment Practitioners (EAPs), and mediocre reports add to the problems associated with effective decision-making. This work presents a framework aimed at improving the effectiveness and objectivity of the decision-making process applied in South Africa’s EIAs in the energy sector. A number of decision-making models and tools are available to researchers and practitioners throughout the world that could potentially be applied in EIAs. Among these are Cost-Benefit Analysis (CBA), Rapid Impact Assessment Matrix (RIAM), and Multi Criteria Analysis (MCA). Each of the tools has its own advantages and disadvantages. With respect to the CBA, its biggest disadvantage is the fact that it requires conversion into monetary terms of all impacts, which is sometimes difficult to achieve. The RIAM, on the other hand, fails to provide a systematic approach to the ranking of alternatives. Both of these issues are addressed by the MCA tools. The MCA framework, furthermore, is universal, transparent, easy to replicate, and does not require a particularly large amount of labour and financial resources to complete. It is, however, subjective, but this shortcoming can be overcome by making the decision process more transparent. The framework proposed in this research paper is based on the Multi Criteria Analysis (MCA) technique that allows the identification of the proposed development's cumulative impact versus the current status of the environment. It then compares possible alternatives, where available, in order to identify the most optimal solution. The proposed solution takes into account the trade-offs between the different impact metrics. The research methodology followed in this paper comprised four steps, namely:<ul><li> Selection of case studies, </li><li> Information collection, </li><li> Framework application and testing and </li><li> Feedback. </li></ul> The development of the framework followed an eight-step approach that is generic for MCA and was tested on two case studies that have already gone through the Environmental Impact Assessment process, i.e. the Open Cycle Gas Turbine (OCGT) plant in the Western Cape and the Concentrating Solar Power (CSP) plant in the Northern Cape. The former was evaluated against the "no-go option", but included a decision tree comprised of impact areas, categories of impacts and dimensions (environmental, social, and economic). The latter included alternatives for four components of the project, but the decision tree comprised only of categories and dimensions. The effectiveness of the framework was verified by testing the results of the case studies against the recommendations proposed in the respective Environmental Impact Reports. In all cases, but one, the results of the framework correlated with the recommendations made by the Environmental Assessment Practitioners in the respective studies. In addition, a workshop with the decision-makers was held to obtain their viewpoints regarding the usefulness of the framework in their decision-making environment. These decision-makers supported the use of the framework in their environment as it offered an integrated and transparent approach to the evaluation of projects and alternatives. They emphasised, however, that the decision-making process was complex and the application of the framework alone would not be able to address all the challenges. The case studies demonstrated that the proposed framework could be successfully applied in the process of undertaking impact assessments in the energy sector. It can be used to determine the trade-offs between impacts and dimensions, while taking into consideration the opinions of specialists and decision-makers when assigning weights. The framework has the ability to clearly illustrate the benefit of introducing mitigation measures and it also indicates an alternative that produces the optimal cumulative impact. In conclusion, the work presented contributes to the new body of knowledge in the field of Environmental Impact Assessment in the energy sector as it will assist authorities in making objective and informed decisions, while ensuring greater transparency in the process. It also opens opportunities for conducting follow-on investigations, such the application of the framework in other sectors of the economy, undertaking a sensitivity analysis to compare the range of scores used in the evaluation of impacts, and investigating the possibility of acquiring input from Interested and Affected Parties (I&APs) and integrating those into the framework. / Dissertation (MSc)--University of Pretoria, 2011. / Graduate School of Technology Management (GSTM) / unrestricted

No description available.