Thesis--Wisconsin. / Vita. Includes bibliographical references (leaves 65-67).

Universiẗat, Diss., 2004--Dortmund.

Zugl.: Freiburg (Breisgau), Pädagog. Hochsch., Diss., 2009.

No description available.

The aim of Master's thesis is to apply statistical methods in production process. The assignment of the thesis is to describe and evaluate pressing process which is the part of posistor production. The first theoretical part contains introduction to the mathematical statistics, verifying assumptions about the data, describing the regression analysis, the analysis of variance and the part of the design of experiment. The second practical part is focused on design, performing and interpreting of experiments. The thesis is developed according to requirements of company EPCOS s.r.o. Šumperk. Statistical program MINITAB Rlease 14, which is supported by company, was used for all computations.

Uniform design has found successful applications in manufacturing, system engineering, pharmaceutics and natural sciences since it appeared in 1980's. Recently, research related to uniform design is emerging. Discussions are mainly focusing on the construction and the theoretical properties of uniform design. On one hand, new construction methods can help researchers to search for uniform designs in more efficient and effective ways. On the other hand, since uniformity has been accepted as an essential criterion for comparing fractional factorial designs, it is interesting to explore its relationship with other criteria, such as aberration, orthogonality, confounding, etc. The first goal of this thesis is to propose new uniform design construction methods and recommend designs with good uniformity. A novel stochastic heuristic technique, the adjusted threshold accepting algorithm, is proposed for searching uniform designs. This algorithm has successfully generated a number of uniform designs, which outperforms the existing uniform design tables in the website https://uic.edu.hk/~isci/UniformDesign/UD%20Tables.html. In addition, designs with good uniformity are recommended for screening either qualitative or quantitative factors via a comprehensive study of symmetric orthogonal designs with 27 runs, 3 levels and 13 factors. These designs are also outstanding under other traditional criteria. The second goal of this thesis is to give an in-depth study of the uniformity property of fractional factorial designs. Close connections between different criteria and lower bounds of the average uniformity have been revealed, which can be used as benchmarks for selecting the best designs. Moreover, we find non-isomorphic designs have different combinatorial and geometric properties in their projected and level permutated designs. Two new non-isomorphic detection methods are proposed and utilized for classifying fractional factorial designs. The new methods take advantages over the existing ones in terms of computation efficiency and classification capability. Finally, the relationship between uniformity and isomorphism of fractional factorial designs has been discussed in detail. We find isomorphic designs may have different geometric structure and propose a new isomorphic identification method. This method significantly reduces the computational complexity of the procedure. A new uniformity criterion, the uniformity pattern, is proposed to evaluate the overall uniformity performance of an isomorphic design set.

Uniform design has found successful applications in manufacturing, system engineering, pharmaceutics and natural sciences since it appeared in 1980's. Recently, research related to uniform design is emerging. Discussions are mainly focusing on the construction and the theoretical properties of uniform design. On one hand, new construction methods can help researchers to search for uniform designs in more efficient and effective ways. On the other hand, since uniformity has been accepted as an essential criterion for comparing fractional factorial designs, it is interesting to explore its relationship with other criteria, such as aberration, orthogonality, confounding, etc. The first goal of this thesis is to propose new uniform design construction methods and recommend designs with good uniformity. A novel stochastic heuristic technique, the adjusted threshold accepting algorithm, is proposed for searching uniform designs. This algorithm has successfully generated a number of uniform designs, which outperforms the existing uniform design tables in the website https://uic.edu.hk/~isci/UniformDesign/UD%20Tables.html. In addition, designs with good uniformity are recommended for screening either qualitative or quantitative factors via a comprehensive study of symmetric orthogonal designs with 27 runs, 3 levels and 13 factors. These designs are also outstanding under other traditional criteria. The second goal of this thesis is to give an in-depth study of the uniformity property of fractional factorial designs. Close connections between different criteria and lower bounds of the average uniformity have been revealed, which can be used as benchmarks for selecting the best designs. Moreover, we find non-isomorphic designs have different combinatorial and geometric properties in their projected and level permutated designs. Two new non-isomorphic detection methods are proposed and utilized for classifying fractional factorial designs. The new methods take advantages over the existing ones in terms of computation efficiency and classification capability. Finally, the relationship between uniformity and isomorphism of fractional factorial designs has been discussed in detail. We find isomorphic designs may have different geometric structure and propose a new isomorphic identification method. This method significantly reduces the computational complexity of the procedure. A new uniformity criterion, the uniformity pattern, is proposed to evaluate the overall uniformity performance of an isomorphic design set.

The aim of this dissertation is to examine selective exposure theory on Twitter among student users at Kuwait University, and to revisit selective exposure theory’s assumptions in the social media era. Two studies for this dissertation have been conducted among a total of 1391 participants to examine the selective exposure theory among student Twitter users. In both studies, the researcher conducted an online experiment by developing simulated Twitter interface pages and a simulated news app to study selective exposure theory among Kuwait University students. The first study aimed to examine whether the students at Kuwait University tend to be exposed to politicians in Kuwait’s parliament who share the same political ideologies. The second study aimed to examine to what extent student users selectively expose themselves to specific content on Twitter, or more specifically to their like-minded group, and what drives them to do so. Moreover, the effect of Twitter’s social endorsement features on users' news selection has been examined. Each study sample has been drawn from different classes of students at Kuwait University. This study also looked at the role of incidental exposure as a means of encouraging cross-ideological exposure. One noticeable trend in the two experiments conducted for this dissertation is that partisan selective exposure was clear among students participating in both studies, but at different levels. Also, data showed that there was no clear role for the social endorsements on Twitter among students in this experiment to reduce selectivity. Moreover, a person's political leaning is more likely to surpass the impact of the social endorsements when users are browsing Twitter on a daily basis. Results showed that students in the second study read tweets from accounts they did not follow in real life and they asserted that they experienced that on a high basis. Implications of these two online experiment studies are discussed.

No description available.

The delivery of immunoreactive macromolecules to tumour cells in solid, heterogeneously perfused tumours is a major problem in the effectiveness of immunotherapy. To help optimize the new experimental treatment method, a published mathematical model of macromolecule transport (Baxter & Jain 1989,1990,199la) was appraised and verified experimentally. Computational and analytical tools were developed to predict the interstitial plasma fluid pressure and velocity distributions in well perfused spherical tumours. Their published analytical solutions of the formulation were found to have some errors and were corrected in this work. To check the validity of the formulation, a series of animal experiments was performed to quantify the total vascular volume, and plasma fluid extravasation rate in SKOV3ipl human ovarian tumour xenografts in nude mice. The results compared well with the theoretically predicted total plasma fluid extravasation rate. Computer codes were also developed to predict the spatial and temporal distributions of intact IgG and its F(ab')₂ and Fab/Fab' fragments in well perfused spherical tumours using the formulation proposed by Baxter & Jain (1989,1990,1991a). The cases of non-binding and binding macromolecules were treated separately. The codes for both the interstitial pressure and macromolecule distributions were written to include a radially variable vessel surface area for transcapillary exchange per unit volume of tumour tissue (SN). The sensitivity of the overall tumour perfusion to variation of (a) the macromolecule m.w., binding affinity, and metabolism, (b) SN, tumour radius, and (c) microvascular permeability were investigated. Comparison of the theoretical predictions with available experimental data leads to the realization of a number of shortcomings in the previously proposed formulations. Finally, a computational method for deriving the effective spherically symmetric spatial distributions for the vascular volume density, and SN from tumour serial sections was developed. This bridges the gap between the actual topology of vascular distributions in tumours and the format of current formulations. / Thesis / Master of Science (MS)