Currently the trend of electronic product development is to ward ¡§light and thin, multi-functional, high density and durability¡¨. When the microelectronic chips tend to be high power, high density and high speed, the rapid increase of heat in a reduced unit area of package size, will lead to failure of electronic products. The contents of thesis is to find out the dominant factors in heat transfer by changing the geometries and material properties of QFN and BGA packages. It also aims to achieve the beat the thermal performance by reducing the probability of failure. In industries it needs a lot of cost and time in experiment work due to the changes of size and materials. Herein, the softwares of ANSYS and ICEPAK are adopted to model the QFN and BGA packages with the statistical experimental design of Taguchi method L18 (21¡Ñ37) orthogonal array setting parameters and obtain the degree of effect for each factor. Eventually, we use the analysis of variance ANOVA to obtain the contribution of each factor and to identify the significant degree for various parameters by variance error integration. From the results the die attach thermal conductivity affects the contribution of thermal performance up to 81.46% for QFN package in comparison with other controlling factors of high significance and high impact effects. Die attach thermal conductivity between 0.5 W/m•k and 1.5 W/m•k the Tj declines much larger than that between 1.5 W/m•k and 8 W/m•k. Die /PKG area ratio affects the contribution of the thermal performance to 64.24% and increasing Die /PKG area ratio can reduce the Tj for BGA package. The significant effect is also higher than other factors. However, the contribution of substrate layers is 18.83% at 99% confidence level.

Cognition, attitudes, behavior and feel things from the environment to explore the small Port of Kaohsiung residents in the air pollution control fee levied in the Republic of China 87 years, for housing the regional air pollution perception of air quality to improve satisfaction and surfacetake the behavior of the current situation of air pollution.To explore the residents of different backgrounds, perception, the air pollution and air quality to improve the satisfaction and take the differences of behavior. This study used a random convenience sampling questionnaire, collection of air pollution fee on air quality questionnaire "to refer to the relevant literature and experts to discuss the small port area residents as a research tool.In this study, questionnaire survey was small Port of Kaohsiung residents to random convenience sampling survey, issued 233 questionnaires were valid questionnaires was 212. Finally, the collected data using descriptive statistics, independent sample t test, one-way ANOVA, Pearson correlation analysis and simple regression and other statistical methods, data processing analysis. The results of this study found that the small port area residents are the upper level air pollution Perception extent. However, a small port area residents as both are not satisfied with the degree of satisfaction for today's small port area of air quality improvement.Acts taken in the face of air pollution is mostly self-protection, and group behavior, and petition the face of lack of government subsidies, while complaints. The results showed that although the high degree of Perception of the small port area residents air pollution, but taken in the face of air pollution, but a means of self-protection, so this study will analyze relevant statistical data and the last suggestions reference, and more availableinformation for the government to improve air quality.

Evaluating the ability of biosorbents to remove arsenic from water has global significance due to the widespread availability and low cost of biosorbent materials. In this study, the ability of coffee grounds and coconut substrate (two previously unreported biosorbents) to remove arsenic from water was compared against the performance of arsenic removal on rice husk (a recognized and widely tested biosorbent). The three biosorbents were individually screened for their ability to remove arsenite, As (III), and arsenate, As (V), from water. Batch reactors were employed to assess the percent removal, reaction kinetics, adsorption capacity, and desorption of each arsenic species onto/from biosorbents under pH buffered and non‐buffered conditions. The resulting experimental data was statistically interpreted using analysis of variance and ttesting of the means. The experimental results were also fit to existing kinetic and isotherm models to provide kinetic rate constants, the maximum adsorption capacity, and to help interpret the nature of the reactions on the biosorbent surface. While all three biosorbents removed arsenic with similar initial reaction kinetics (pseudo 1st order reaction rate constant for As (III) was 0.13 hr^‐1 for all three biosorbents and for As (V) was 0.17 hr^‐1 for coffee grounds and rice husk and 0.15 hr^‐1 for coconut substrate), the amount of arsenite and arsenate removed was highest for coffee grounds (84 and 91 percent, respectively), followed by rice husk (68 and 72 percent, respectively), and then coconut substrate (26 and 24 percent, respectively). The maximum adsorption capacity of arsenite and arsenate was determined for coffee grounds (0.66 and 0.70 mg/g, respectively) and rice husk (0.55 and 0.66 mg/g, respectively). While desorption was observed for both coffee grounds and rice husk, the total amount of desorption accounted for less than 15 percent of the total retained mass. The results of this thesis work reveal that coffee can be used as an effective biosorbent when compared to rice husk; however, coconut substrate is less effective than rice husk at removing As (III) and As (V).

ABSTRACT This thesis gathered the actual Burn-In (BI) data from one of the leading cooperation in the semiconductor industry, and analyzed the major factors¡¦ impact on BGA package solder ball deformation. The Taguchi Method was used for these analyses, and the commercial statistic software MiniTab14 was widely used on this thesis. The solder ball stress was analized by using the commercial FEM software Ansys 8.1. Some electrical characters (such as device power) can be only observed from Burn-In process, but not static acceleration tests. These effects were fully discussed in this thesis. The analyses got the result that the smaller solder ball pitch/solder ball diameter causes the more serious solder ball deformation under the specific socket vendor precondition. Burn-In time are also a significant factor for solder deformation. Basically the longer BI time cause the more serious solder deformation. The device power effect is not significant within the power sampling range of this thesis.

The importance of value lies in its potential influence of people¡¦s behavior and decision. It also determines how people distribute their available resources. After the World War II, in the western developed countries people¡¦s value has changed to post-materialism due to the stable growth of economy in these countries. People transferred their value priority from economy growth, stable job, authority, order, and rationality, to freedom of speech, environmental protection, multi-value, leisure time, art, and self- expression. These changes have caused governments to react these issues think about how to govern the diversity and complicity of changing value. This research focuses on value of time and space transformation by analysing data from Taiwan Social Change Survey (TSCS). Through re-construction scale and calculate score of post-material value, the paper examines trend of value changes from 1985 to 2005 by adopting one-way ANOVA, two-way ANOVA and three-way ANOVA approaches to analyse the differences of value among socio-economic variables. From the findings this research reveals that: (1) In Taiwan, value including period effect, age effect and cohort effect has significant changes along with time passing by; (2) Variables influencing value changes include age, income, education, job, marriage, and urban attributes. Income interacts with urban attribute, and moreover, income, urban attribute, and marriage interact with one another; (3) The effect of gender has not been proven for influencing value change; however, value of female gradually changes toward post-material value; (4) The influence of income value decreases when income decrease; however, this influence does not have significant change between people of age 40 to 59; (5) Job does not have significant influence on secondary industry and tertiary industry in early stages, and it also does not have significant influence on primary industry and secondary industry in lately stages; (6) People with high income, high education, or in unmarried status have high post-material value; (7) The difference of urban attribute shows that there is significant difference between urban, town, and rural, due to urban value changes toward post-material in early stage; however in lately stage there is no significant difference between urban and town. From the finding above it indicates that value changes from urban to town in space respect; (8) Between urban and income interaction of value change, people in urban area show significant difference of income value, then comes to people in town, finally people in rural area. As to the interaction among urban, income and marriage, it shows that single people with low income show significant difference in urban attribute; (9) Governments of urban area with post-material features should distribute resource in post-material governance. This research finds that there are significant post-material features in urban area, and significant material features in rural area; (10) Form indictor of macroeconomics, personal expenditure, interpellation of city council, election bulletin, final accounting of expenditure, this research shows that urban governance also shifts as researched value changes; (11) Since 2000, Taiwan has faced economic recession and raising unemployment problems which cause value changes toward material value. This change leads local governments to decrease expenditure on both environmental protection and community development during 2000 to 2007; (12) Urban government should focus on good governance, and value changing toward post-material contributes to foundation of good governance. Governments should improve the way of governance based on these value changes.

<p>The thesis is dealing with mixing processes havingapplication to ground improvement by deep mixing. The mainobjectives of the thesis is to make a contribution to knowledgeof the basic mechanisms in mixing binding agents into soil andimprove the knowledge concerning factors that influence theuniformity of stabilised soil.</p><p>A great part of the work consists of a literature surveywith particular emphasis on literature on the processindustries. This review forms a basis for a profounddescription and discussion of the mixing process and factorsaffecting the process in connection with deep mixingmethods.</p><p>The thesis presents a method for a simple field test for thestudy of influential factors in the mixing process. A number offactors in the installation process of lime-cement columns havebeen studied in two field tests using statistical multifactorexperiment design. The effects of retrieval rate, number ofmixing blades, rotation speed, air pressure in the storagetank, and diameter of the binder outlet on the stabilisationeffect and the coefficient of variation determined byhand-operated penetrometer tests for excavated lime-cementcolumns, were studied.</p><p>The literature review, the description of the mixingprocess, and the results from the field tests provide a morebalanced picture of the mixing process and are expected to beuseful in connection to ground improvement projects and thedevelopment of mixing equipments.</p><p>The concept of sufficient mixture quality, i.e. theinteraction between the mixing process and the mechanicalsystem, is discussed in the last section. By means ofgeostatistical methods, the analysis considers thevolume-variability relationship with reference to strengthproperties. According to the analysis, the design values forstrength properties depends on the mechanical system, the scaleof scrutiny, the spatial correlation structure, and the conceptof safety, i.e. the concept of sufficient mixture quality isproblem specific.</p><p><b>Key words:</b>Deep Mixing, Lime cement columns, Mixingmechanisms, Mixture quality, Field test, ANOVA, Variancereduction.</p>

Le thème général de cette thèse est celui de la construction de modèles permettant d'approximer une fonction f lorsque la valeur de f(x) est connue pour un certain nombre de points x. Les modèles considérés ici, souvent appelés modèles de krigeage, peuvent être abordés suivant deux points de vue : celui de l'approximation dans les espaces de Hilbert à noyaux reproduisants ou celui du conditionnement de processus gaussiens. Lorsque l'on souhaite modéliser une fonction dépendant d'une dizaine de variables, le nombre de points nécessaires pour la construction du modèle devient très important et les modèles obtenus sont difficilement interprétables. A partir de ce constat, nous avons cherché à construire des modèles simplifiés en travaillant sur un objet clef des modèles de krigeage : le noyau. Plus précisement, les approches suivantes sont étudiées : l'utilisation de noyaux additifs pour la construction de modèles additifs et la décomposition des noyaux usuels en sous-noyaux pour la construction de modèles parcimonieux. Pour finir, nous proposons une classe de noyaux qui est naturellement adaptée à la représentation ANOVA des modèles associés et à l'analyse de sensibilité globale.

Dans un modèle qui peut s'avérer complexe et fortement non linéaire, les paramètres d'entrée, parfois en très grand nombre, peuvent être à l'origine d'une importante variabilité de la sortie. L'analyse de sensibilité globale est une approche stochastique permettant de repérer les principales sources d'incertitude du modèle, c'est-à-dire d'identifier et de hiérarchiser les variables d'entrée les plus influentes. De cette manière, il est possible de réduire la dimension d'un problème, et de diminuer l'incertitude des entrées. Les indices de Sobol, dont la construction repose sur une décomposition de la variance globale du modèle, sont des mesures très fréquemment utilisées pour atteindre de tels objectifs. Néanmoins, ces indices se basent sur la décomposition fonctionnelle de la sortie, aussi connue sous le nom de décomposition de Hoeffding. Mais cette décomposition n'est unique que si les variables d'entrée sont supposées indépendantes. Dans cette thèse, nous nous intéressons à l'extension des indices de Sobol pour des modèles à variables d'entrée dépendantes. Dans un premier temps, nous proposons une généralisation de la décomposition de Hoeffding au cas où la forme de la distribution des entrées est plus générale qu'une distribution produit. De cette décomposition généralisée aux contraintes d'orthogonalité spécifiques, il en découle la construction d'indices de sensibilité généralisés capable de mesurer la variabilité d'un ou plusieurs facteurs corrélés dans le modèle. Dans un second temps, nous proposons deux méthodes d'estimation de ces indices. La première est adaptée à des modèles à entrées dépendantes par paires. Elle repose sur la résolution numérique d'un système linéaire fonctionnel qui met en jeu des opérateurs de projection. La seconde méthode, qui peut s'appliquer à des modèles beaucoup plus généraux, repose sur la construction récursive d'un système de fonctions qui satisfont les contraintes d'orthogonalité liées à la décomposition généralisée. En parallèle, nous mettons en pratique ces méthodes sur différents cas tests.

Immer mehr Eltern wollen zur Sicherung ihres Lebensstandards auch in den ersten Lebensjahren ihrer Kinder in Vollzeit berufstätig werden. Deshalb nimmt der Aufenthalt in Kindertageseinrichtungen und Schulen einen Großteil des Tages ein. Bereits seit 1996 besteht für jedes Kind im Alter vom vollendeten dritten Lebensjahr bis zum Beginn der Grundschule ein Rechtsanspruch auf einen Platz in einer Kindertagesstätte. Seit August 2013 beginnt dieser Anspruch bereits mit dem vollendeten ersten Lebensjahr. (vgl. Textor, 2010) Die Anzahl der ganztägig geöffneten Kindertageseinrichtungen nimmt stetig zu und immer mehr Schulen bieten Nachmittagsbetreuung an bzw. wandeln sich in Ganztagsschulen um. Aufgrund der steigenden Inanspruchnahme der Angebote der Kindertagesstätten rückt die Familienbetreuung immer weiter in den Hintergrund. (vgl. Textor, 2010) Somit gewinnt jedoch die Frage nach der richtigen Betreuungsmöglichkeit für das eigene Kind immer mehr an Bedeutung. Hierbei reicht das vielfältige Angebot von den reformpädagogischen Ansätzen Friedrich Fröbels, Maria Montessoris oder Rudolf Steiners bis hin zu den modernen Ansätzen wie der offene Kindergarten, der Situationsansatz oder die Reggio-Pädagogik. (vgl. Knauf, 2010, S. 222 & 228) Dadurch wird die Entscheidung, welcher Ansatz für das eigene Kind und dessen optimalen Entwicklung auch für die spätere berufliche Zukunft am sinnvollsten ist, erschwert. Im Rahmen dieser Arbeit soll versucht werden, einen Ansatz zur einfacheren Entscheidung zu geben. In dem eine Gegenüberstellung der Montessori-Pädagogik, Waldorf-Pädagogik, Reggio-Pädagogik und dem Situationsansatz im Kindergartenalter auf ausgewählte Kompetenzen vorgenommen wird. Dazu zählen die naturwissenschaftliche Kompetenz sowie das Hörverstehen auf Satz- und Wortebene. Signifikante Unterschiede der dargestellten Konzepte und Kompetenzen sollen mittels Varianzanalyse identifiziert werden. / The comparison of the Montessori pedagogy, the Waldorf pedagogy, the Reggio pedagogy and the situational approach at kindergarten age on scientific competence, listening comprehension at sentence level and word level as well as procedurale metacognition. Identification of significant connections between the concepts and selected competencies.

L'objectif de cette thèse est l'investigation de nouvelles méthodes de surface de réponse afin de réaliser l'analyse de sensibilité de modèles numériques complexes et coûteux en temps de calcul. Pour ce faire, nous nous sommes intéressés aux méthodes basées sur la décomposition ANOVA. Nous avons proposé l'utilisation d'une méthode basée sur les splines de lissage de type ANOVA, alliant procédures d'estimation et de sélection de variables. L'étape de sélection de variable peut devenir très coûteuse en temps de calcul, particulièrement dans le cas d'un grand nombre de paramètre d'entrée. Pour cela nous avons développé un algorithme de seuillage itératif dont l'originalité réside dans sa simplicité d'implémentation et son efficacité. Nous avons ensuite proposé une méthode directe pour estimer les indices de sensibilité. En s'inspirant de cette méthode de surface de réponse, nous avons développé par la suite une méthode adaptée à l'approximation de modèles très irréguliers et discontinus, qui utilise une base d'ondelettes. Ce type de méthode a pour propriété une approche multi-résolution permettant ainsi une meilleure approximation des fonctions à forte irrégularité ou ayant des discontinuités. Enfin, nous nous sommes penchés sur le cas où les sorties du simulateur sont des séries temporelles. Pour ce faire, nous avons développé une méthodologie alliant la méthode de surface de réponse à base de spline de lissage avec une décomposition en ondelettes. Afin d'apprécier l'efficacité des méthodes proposées, des résultats sur des fonctions analytiques ainsi que sur des cas d'ingénierie de réservoir sont présentées.