Cette thèse se penche sur les problèmes de décisions devant être prises de manière séquentielle au sein d'un environnement aléatoire. Lors de chaque étape d'un tel problème décisionnel, une alternative doit être sélectionnée parmi un ensemble d'alternatives. Chaque alternative possède un gain moyen qui lui est propre et lorsque l'une d'elles est sélectionnée, celle-ci engendre un gain aléatoire. La sélection opérée peut suivre deux types d'objectifs.<p>Dans un premier cas, les tests viseront à maximiser la somme des gains collectés. Un juste compromis doit alors être trouvé entre l'exploitation et l'exploration. Ce problème est couramment dénommé dans la littérature scientifique "multi-armed bandit problem".<p>Dans un second cas, un nombre de sélections maximal est imposé et l'objectif consistera à répartir ces sélections de façon à augmenter les chances de trouver l'alternative présentant le gain moyen le plus élevé. Ce deuxième problème est couramment repris dans la littérature scientifique sous l'appellation "selecting the best".<p>La sélection de type gloutonne joue un rôle important dans la résolution de ces problèmes de décision et opère en choisissant l'alternative qui s'est jusqu'ici montrée optimale. Or, la nature généralement aléatoire de l'environnement rend incertains les résultats d'une telle sélection. <p>Dans cette thèse, nous introduisons une nouvelle quantité, appelée le "gain espéré d'une action gloutonne". Sur base de quelques propriétés de cette quantité, de nouveaux algorithmes permettant de résoudre les deux problèmes décisionnels précités seront proposés.<p>Une attention particulière sera ici prêtée à l'application des techniques présentées au domaine de la sélection de modèles en l'apprentissage artificiel supervisé. <p>La collaboration avec le service d'anesthésie de l'Hôpital Erasme nous a permis d'appliquer les algorithmes proposés à des données réelles, provenant du milieu médical. Nous avons également développé un système d'aide à la décision dont un prototype a déjà été testé en conditions réelles sur un échantillon restreint de patients. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

La moitié des forêts tropicales du monde est désignée comme des forêts de production par les services forestiers nationaux. Si l'effet de la déforestation sur les émissions de carbone et les pertes de biodiversité en région tropicale ont fait l'objet de nombreuses études, les impacts à long terme de l'exploitation commerciale de bois sur ces forêts restent à ce jour très peu étudiés. Or, les enjeux sont grandissants pour ces forêts de production. Non seulement elles doivent alimenter le commerce du bois tropical, mais leur rôle dans le stockage de carbone et dans la préservation de la biodiversité est de plus en plus reconnu. La thèse développée ici s'organise en trois temps. Premièrement, un modèle de bilan carbone de l'exploitation forestière est développé, et les différences régionales de dynamique de récupération du carbone post-exploitation sont modelisées. Deuxièmement, un modèle de récupération du volume de bois d'oeuvre a été développé et calibré à l'échelle amazonienne. Ces résultats ont permis de montrer la lenteur de la récupération du volume de bois d'Amazonie, et la non-durabilité des pratiques actuelles d'exploitation. Ces modèles (carbone et bois d'oeuvre) ont été élaborés dans un cadre bayésien, avec l'appui et les données du Tropical managed Forest Observatory (www.tmfo.org), réseau rassemblant 9 institutions de recherche et plus de 200 parcelles de suivi de la dynamique forestière après exploitation en Amazonie. Enfin, une analyse comparative de scénarios prospectifs a été effectuée, où les compromis possibles entre services écosystémiques (bois d'oeuvre, carbone et biodiversité) ont été explorés par des techniques d'optimisation multicritères. / Half of the world's tropical forets are designated by national forest services as production forests. While here have been numerous studies on the impact if devorestation on carbon emissions and biodiversity loss in the tropics, the long-term impacts of selective logging are still poorly studied; However, the importance of these production forests is increasing. Not only must they meet the growing tropical timber demand, but their role in Carbon storage and the biodiversity conservation is increasingly recognized. The studies developped in this thesis are threefold. First, a carbon balance model of logging is developped, and regional differences in post-logging carbon recovery of timber volume on the Amazon, and thus the unsustainability of current logging practices. These models (carbon and timber) were developped on a Bayesian framework, with support and data from the Tropical managed Forest Observatory (TmFO : www.tmfo.org), a netword of 9 research institutions and more than 200 forest plotswhere post-logging forest dynamics have been monitored for <30 years in the Amazon. Finally, a comparative analysis of prospective scenarios was carried out, where potential tradeoffs between ecosystem services (timber, carbone and biodiversity) were explored using multi-criteria optimization analysis.

Text in English / The question the researcher set out to answer during this research study was: "What is the Educational Focus of a Nursing College, when viewed within Bevis and Watson's Humanistic·Educative-Caring Curriculum Paradigm versus a Stimulus-Response Curriculum Paradigm?" To answer this question, an exploratory, descriptive and contextual study was undertaken. The population consisted of tutors employed at a College of Nursing. From this population a purposive sample was drawn. Data was collected by means of formal, semi-structured interviews and a questionnaire. Content analysis was used to analyse the data. It was found that during the first year of students' education and training, the Educational Focus of the College was on training but during the second year, the focus changed to the transitional phase. The transitional phase is situated between training and education. During the third and fourth year, the focus was also in transition, but the way students learn was at the training phase. / Health Studies / M.A. (Advanced Nursing Sciences)

由近期的研究中發現變動所有參數的管制圖在偵測小幅度偏移時的速度比起傳統的舒華特管制圖來的快，許多文獻也討論到利用調適性管制技術同時監控製程的平均數和變異數。而在這份研究中，為了改善現有管制圖的偵測效率，依序提出了U-V管制圖以及Max-M管制圖來偵測單一製程與兩相依製程的平均數和變異數。採用AATS及ANOS來衡量管制圖的偵測績效，並利用馬可夫鏈推導計算得之。透過兩階段的範例來介紹所提出的管制圖的應用方法並將VP U-V管制圖、VP Max-M管制圖與FP Z(X-bar)-Z(Sx^2)管制圖加以比較。從所研究的數值分析中發現VP Max-M管制圖比另兩種管制圖的表現來的好，再加上只需要單一管制圖在使用上對工程師來說也較為簡便，因此建議Max-M管制圖値得在實務上被使用。 / Recent studies have shown that the variable parameters (VP) charts detect small process shifts faster than the traditional Shewhart charts. There have been many papers discussed adaptive control schemes to monitor process mean and variance simultaneously. In the study, to improve the efficiency and performance of the existing control charts, the U-V control charts and Max-M control charts are respectively proposed to monitor the process mean and variance for a single process and two dependent process steps. The performance of the proposed control charts is measured by using adjusted average time to signal (AATS) and average number of observations to signal (ANOS). The calculation of AATS and ANOS is derived by Markov chain approach. The application of the proposed control charts is illustrated by a numerical example for two dependent process steps, and the performance of VP U-V control charts, VP Max-M control charts and FP Z(X-bar)-Z(Sx^2) control charts is compared. From the results of data analyses, it shows that the VP Max-M control charts have better performance than VP U-V control charts and FP Z(X-bar)-Z(Sx^2) control charts. Furthermore, using a single chart to monitor a process is easier than using two charts for engineers. Hence, Max-M control charts are recommended in real industrial process.

市場比較法估價過程需要經比較、分析及調整三階段，而估價師於個別估價應用中因缺乏標準化依據，造成估價過程常被質疑過於主觀且偏重經驗法則，導致估價結果產生因人而異現象。基於此本研究乃建立大量估價模型，運用估價師進行市場比較法行為邏輯，在選取比較案例階段採用明科斯基距離概念，並結合特徵價格理論，建立大量樣本的客觀標準化估價模型。藉由此模型分析，估價師不需要主觀預測，可改善過去估價結果不一致情形，並達到大量估價目的。 透過隨機抽樣的重覆實驗，實證結果發現，以模型的平均百分比預測誤差與命中率比較，整體而言未劃分次市場估價模型皆較劃分次市場準確，且其穩定度亦較高，而劃分次市場較不準確的原因，在於舊市區的表現不穩定，與市郊區的估值表現較差所致。經本研究結果，不論劃分次市場與否，標準化大量估價模型大致上均達到研究預設水準，然劃分次市場模型雖有時較準確但未優於未分區模型甚多，顯示適當劃分次市場可提高準確度，但資料太少時，劃分次市場可能產生統計偏誤問題。 / The process of the market comparative approach includes three stages：comparing ,analyzing and adjusting. Real estate appraisers lack the standardized basis in the individual appraisal application, so they are often challenged by subjectivity and stressing experience, which leads to the phenomenon that appraisal results are always different from person to person. Based on this, our research establishes the automated valuation methods. By applying the appraisers’ behavior in market comparative approach, we use minkowski metric in selecting comparative subject, and associate with hedonic price theory to establish objective standardized real estate appraisal model. By using this model, the appraisers can avoid subjective forecasting, reduce inconsistency, and therefore achieve the goal of mass appraisal. Through the repetition experiment of random sampling, we compare MAPE and Hit-rate between models. The result shows that non-delimited markets are more accurate and stable than delimited markets. The reason for this is the instability caused by older district of the city and bad performance in the suburban area. After these findings, no matter delimited market or not, the standardized mass appraisal model achieved the research standard we had set in advance. Sometimes the model of delimited market is more accurate, though not by a significant amount, it shows that higher accuracy can be expected through adequate market segmentation, but will lead to bias when lack of datum.

Text in English / The question the researcher set out to answer during this research study was: "What is the Educational Focus of a Nursing College, when viewed within Bevis and Watson's Humanistic·Educative-Caring Curriculum Paradigm versus a Stimulus-Response Curriculum Paradigm?" To answer this question, an exploratory, descriptive and contextual study was undertaken. The population consisted of tutors employed at a College of Nursing. From this population a purposive sample was drawn. Data was collected by means of formal, semi-structured interviews and a questionnaire. Content analysis was used to analyse the data. It was found that during the first year of students' education and training, the Educational Focus of the College was on training but during the second year, the focus changed to the transitional phase. The transitional phase is situated between training and education. During the third and fourth year, the focus was also in transition, but the way students learn was at the training phase. / Health Studies / M.A. (Advanced Nursing Sciences)

This work is aimed at the problems of finding and selecting employees through the application of the latest knowledge and approach of staff selection. From the knowledge gained it suggests a possible method of choosing employees in the company, ABC Ltd. for the position of a dealer.

Abstract The focus of the thesis is researching how to optimize a certain route, highlighting theimportance of weather conditions in the context of optimizing truck routes. Additionally, itemphasizes the goals of maximizing human utilization and cost-effectiveness. The project teamconducted a survey on fuel habits, and the results showed that the participating companies lackeda defined strategy for the company's fuel habits or a specific approach for selecting fuelingcompanies. The project team developed an algorithm that assists drivers in selecting the mostoptimized routes while enabling companies to calculate these routes efficiently. Data wascollected on the fuel consumption associated with the loading and unloading of goods. The totalroute cost is calculated using a custom equation that includes the driver´s salary, the employer´ssocial security contribution, the cost of the truck when loading and unloading, and the wear onthe truck. The company naturally expects to generate a profit, and such profit margins may differamong companies. Therefore, the project team excluded this factor from the equation todetermine the route's cost. The findings showed that varying weather conditions did impactconsumption, thereby influencing the final cost. The algorithm that the project team made is asignificant milestone toward the completion of the final product. However, there are still a fewremaining steps to be taken in order to achieve the desired end result and future development.

The question the researcher set out to answer was "What is the educational focus of a nursing college when viewed within Bevis and Watson's Humanistic-Educative-Caring Curriculum Paradigm versus a Stimulus-Response Curriculum Paradigm?" The purpose of this study was to develop and test an instrument based on the Bevis and Watson Humanistic-Educative-Caring Model; an educational paradigm shift from the Tylerian rationale in nursing education. A questionnaire comprising 181 Two-Choice Comparative-Value-Statement Items was developed and tested. A non-experimental research design was implemented. During the developmental phase, a non-probability, purposive sample was used; the questionnaire (instrument) was developed; data were analysed by applying content analysis and the questionnaire was refined. During the testing phase a stratified, random sample was used consisting of first to fourth year students from two nursing colleges from the Gauteng Province; the items were tested against biographic data and hypotheses resulting from the Bevis and Watson model. Six conceptual continuums comprise the Bevis and Watson model namely, the Learner Maturity Continuum, the Teacher-student relationship, the Teacher-student structure, the Typology of Learning, Criteria for Teacher-Student Interactions and Criteria for Selecting and Devising Learning Experiences. Both descriptive and inferential statistics were utilised. The results indicated that the educational focus of the respondents with regard to the Bevis and Watson model was predominantly humanistic. The exception occurred with regard to TUTOR: Teacher-student structure; pertaining to hypothesis 7 on language; and hypothesis 9 on gender, where a behaviouristic orientation appears to prevail. Significant differences were found between the model variables (conceptual continuums) and year group, language, college A and B, and gender. In hypothesis 5, although a humanistic orientation predominated, the 4th year students tended to display an increasing behaviouristic orientation. In hypothesis 8, although a humanistic orientation predominated, college A appeared less humanistic than college B. Recommendations were made regarding nursing education and further research studies to refine the instrument. The implementation of the Bevis and Watson model calls for a curriculum paradigm shift in nursing education. / Health Studies / D.Litt. et Phil.

In South Africa household debt has increased rapidly over the past few years, therefore illustrating the importance of analysing liability usage behaviour of individual members within the household. In order to comprehend the behaviour of South Africans regarding liability usages, this study provides insight into why individuals find it necessary to obtain liability products as well as to indicate whether liability products are used to address the financial needs for the purpose it was developed for. To achieve the aim of this study, it was firstly necessary to develop a theoretical framework for the process of selecting credit products when satisfying financial needs. Secondly, the characteristics and intended usage purposes of different credit products available in South Africa were discussed and a debt classification framework was developed. Finally, data obtained from the Finscope South Africa survey was analysed according to the developed frameworks following a combination of two approaches. Firstly, a qualitative approach was used to identify the different financial needs which are satisfied when using liabilities. The financial needs identified were classified according to Alderfer’s existence relatedness growth (ERG) theory and the factors that have an influence on liability usage. Secondly, a quantitative approach was followed to indicate which financial needs are fulfilled when using different credit products. The results of this study suggest that individuals do not use liabilities only for the purpose what the products were originally developed for. The findings clearly indicated that individuals mainly use liabilities to satisfy basic needs which are classified as existence needs according to Alderfer’s ERG theory. Based on the data analysis a variety of factors such as access to credit and certain demographic characteristics have an influence on liability usage behaviour of individuals. The results further show that individuals mainly use informal, unsecured, short-term loans when satisfying their financial needs which might indicate that South Africans are unable to access formal credit products. / Financial Accounting / M. Phil. (Accounting Science)