Shorter product life cycles and rapid product obsolescence provide increasing incentives to introduce newproducts to markets more quickly. As a consequence of rapidly changing market conditions, firms focus onimproving their new product development processes to reap the benefits of early market entry. Researchershave analyzed market entry, but have seldom provided quantitative approaches for the product rolloverproblem. This research builds upon the literature by using established optimization methods to examine howfirms can minimize their net loss during the rollover process. Specifically, our work explicitly optimizes thetiming of removal of old products and introduction of new products, the optimal strategy, and the magnitudeof net losses when the market entry approval date of a new product is unknown. In the first paper, we use theconditional value at risk to optimize the net loss and investigate the effect of risk perception of the manageron the rollover process. We compare it to the minimization of the classical expected net loss. We deriveconditions for optimality and unique closed-form solutions for single and dual rollover cases. In the secondpaper, we investigate the rollover problem, but for a time-dependent demand rate for the second producttrying to approximate the Bass Model. Finally, in the third paper, we apply the data-driven optimizationapproach to the product rollover problem where the probability distribution of the approval date is unknown.We rather have historical observations of approval dates. We develop the optimal times of rollover and showthe superiority of the data-driven method over the conditional value at risk in case where it is difficult to guessthe real probability distribution

In Operations Enduring Freedom and Iraqi Freedom, explosions accounted for 81% of all injuries; this is a higher casualty percentage than in any previous wars. Blast wave overpressure has recently been associated with varying levels of traumatic brain injury in soldiers exposed to blast loading. Presently, the injury mechanism behind primary blast brain injury is not well understood due to the complex interactions between the blast wave and the human body. Despite these limitations in the understanding of head injury thresholds, head kinematics are often used to predict the overall potential for head injury. The purpose of this study was to investigate head kinematics, and predict injury from a range of simulated blast loads at varying standoff distances and differing heights of bursts. The validated Generator of body data multi-body human surrogate model allows for numerical kinematic data simulation in explicit finite element method fluid structure interaction blast modeling. Two finite element methods were investigated to simulate blast interaction with humans, an enhanced blast uncoupled method, and an Arbitrary Lagrangian Eularian fully coupled method. The enhanced blast method defines an air blast function through the application of a blast pressure wave, including ground reflections, based on the explosives relative location to a target; the pressures curves are based on the Convention Weapons databases. LBE model is efficient for parametric numerical studies of blast interaction where the target response is the only necessary result. The ALE model, unlike classical Lagrangian methods, has a fixed finite element mesh that allows material to flow through it; this enables simulation of large deformation problems such as blast in an air medium and its subsequent interaction with structures. The ALE model should be used when research into a specific blast scenario is of interest, since this method is more computationally expensive. The ALE method can evaluate a blast scenario in more detail including: explosive detonation, blast wave development and propagation, near-field fireball effects, blast wave reflection, as well as 3D blast wave interaction, reflection and refraction with a target. Both approaches were validated against experimental blast tests performed by Defense Research and Development Valcartier and ConWep databases for peak pressure, arrival time, impulse, and curve shape. The models were in good agreement with one another and follow the experimental data trend showing an exponential reduction in peak acceleration with increasing standoff distance until the Mach stem effect reached head height. The Mach stem phenomenon is a shock front formed by the merging of the incident and reflected shock waves; it increases the applied peak pressure and duration of a blast wave thus expanding the potential head injury zone surrounding a raised explosive. The enhanced blast model was in good agreement with experimental data in the near-field, and mid-field; however, overestimated the peak acceleration, and head injury criteria values in the far-field due to an over predicted pressure impulse force. The ALE model also over predicted the response based on the head injury criteria at an increased standoff distance due to smearing of the blast wave over several finite elements leading to an increased duration loading. According to the Abbreviated Injury Scale, the models predicted a maximal level 6 injury for all explosive sizes in the near-field, with a rapid acceleration of the head over approximately 1 ms. There is a drastic exponential reduction in the insult force and potential injury received with increasing standoff distance outside of the near-field region of an explosive charge.

In this dissertation, generalized additive neural networks (GANNs) and multilayer perceptrons (MLPs) are studied and compared as prediction techniques. MLPs are the most widely used type of artificial neural network (ANN), but are considered black boxes with regard to interpretability. There is currently no simple a priori method to determine the number of hidden neurons in each of the hidden layers of ANNs. Guidelines exist that are either heuristic or based on simulations that are derived from limited experiments. A modified version of the neural network construction with cross–validation samples (N2C2S) algorithm is therefore implemented and utilized to construct good MLP models. This algorithm enables the comparison with GANN models. GANNs are a relatively new type of ANN, based on the generalized additive model. The architecture of a GANN is less complex compared to MLPs and results can be interpreted with a graphical method, called the partial residual plot. A GANN consists of an input layer where each of the input nodes has its own MLP with one hidden layer. Originally, GANNs were constructed by interpreting partial residual plots. This method is time consuming and subjective, which may lead to the creation of suboptimal models. Consequently, an automated construction algorithm for GANNs was created and implemented in the SAS R statistical language. This system was called AutoGANN and is used to create good GANN models. A number of experiments are conducted on five publicly available data sets to gain insight into the similarities and differences between GANN and MLP models. The data sets include regression and classification tasks. In–sample model selection with the SBC model selection criterion and out–of–sample model selection with the average validation error as model selection criterion are performed. The models created are compared in terms of predictive accuracy, model complexity, comprehensibility, ease of construction and utility. The results show that the choice of model is highly dependent on the problem, as no single model always outperforms the other in terms of predictive accuracy. GANNs may be suggested for problems where interpretability of the results is important. The time taken to construct good MLP models by the modified N2C2S algorithm may be shorter than the time to build good GANN models by the automated construction algorithm / Thesis (M.Sc. (Computer Science))--North-West University, Potchefstroom Campus, 2011.

In this dissertation, generalized additive neural networks (GANNs) and multilayer perceptrons (MLPs) are studied and compared as prediction techniques. MLPs are the most widely used type of artificial neural network (ANN), but are considered black boxes with regard to interpretability. There is currently no simple a priori method to determine the number of hidden neurons in each of the hidden layers of ANNs. Guidelines exist that are either heuristic or based on simulations that are derived from limited experiments. A modified version of the neural network construction with cross–validation samples (N2C2S) algorithm is therefore implemented and utilized to construct good MLP models. This algorithm enables the comparison with GANN models. GANNs are a relatively new type of ANN, based on the generalized additive model. The architecture of a GANN is less complex compared to MLPs and results can be interpreted with a graphical method, called the partial residual plot. A GANN consists of an input layer where each of the input nodes has its own MLP with one hidden layer. Originally, GANNs were constructed by interpreting partial residual plots. This method is time consuming and subjective, which may lead to the creation of suboptimal models. Consequently, an automated construction algorithm for GANNs was created and implemented in the SAS R statistical language. This system was called AutoGANN and is used to create good GANN models. A number of experiments are conducted on five publicly available data sets to gain insight into the similarities and differences between GANN and MLP models. The data sets include regression and classification tasks. In–sample model selection with the SBC model selection criterion and out–of–sample model selection with the average validation error as model selection criterion are performed. The models created are compared in terms of predictive accuracy, model complexity, comprehensibility, ease of construction and utility. The results show that the choice of model is highly dependent on the problem, as no single model always outperforms the other in terms of predictive accuracy. GANNs may be suggested for problems where interpretability of the results is important. The time taken to construct good MLP models by the modified N2C2S algorithm may be shorter than the time to build good GANN models by the automated construction algorithm / Thesis (M.Sc. (Computer Science))--North-West University, Potchefstroom Campus, 2011.

This research focuses on educators' experience of implementing outcomes-based education in grade nine in secondary schools in South Africa. Two schools were chosen as settings for the qualitative research project. Three focus groups with 20 educators, two focus groups with 14 managers and seven personal interviews were conducted. Twelve classrooms were also observed where grade nine learners were being taught. Findings indicated that some educators found the experience of implementing OBE positive in that it improved their repertoire of facilitating and assessing skills. The reasons for citing OBE as a negative experience were given as the following: an increased workload, poor training and lack of follow-up by the Department and the school management team's degree of involvement. Recommendations were made on how to ease the burden on educators implementing OBE in their classrooms and to empower school managers to manage the implementation of OBE within their respective areas of responsibility. / Educational Studies / M. Ed. (Education Management)

The democratic era in South Africa has led to the introduction of a wide-ranging series of teaching and learning policies aimed at school reform. The study aimed at establishing how these policies are implemented by educators in selected schools. Qualitative and quantitative methods were used to gather information from educators, members of the School Management Teams and learners through in-depth individual and focus group interviews as the main data collection methods. Critical discourse analysis was used to examine participants‟ spoken words and the content of relevant documents such as lesson plans in detail. Comparisons were drawn and similarities identified through the coding, categorisation and condensation of data. The researcher interpreted the displayed data. A discussion of the main themes was presented and supported by quotations by participants. Findings indicate that there have been only few changes in terms of how educators teach; some educators conceded that they had not changed the way they teach since the introduction of new curriculum and assessment policies. The role played by learning outcomes in teachers‟ lesson planning was uneven. Members of School Management Teams demonstrated diverse views on teaching and learning policies. Learners who are taught in their home language encounter few difficulties in learning; this is not the case with learners who are taught through medium of a second language and thus lack the necessary language proficiency necessary for academic achievement. A shift from teacher-centred to learner-centred approaches to teaching is still required: that is, a shift from teaching to learning to enhance thinking and reflection which accommodates and draws on learners‟ prior knowledge and experiences. Class sizes should be reduced to enable effective educator and learner interaction. Educators‟ interest, tolerance, innovativeness and competency in the subject matter will help sustain learners‟ interest in the learning activities. Policy makers should explore professional development from viewpoint of the participating educators in order to identify the most effective strategies to support and change educators‟ classroom practice where necessary. / Curriculum and Instructional Studies / D. Ed. (Curriculum Studies)

Text in Afrikaans, abstract in Afrikaans and English / Navorsing toon dat beoordelingskale, gebaseer op spesifieke werksgedrag as geldige prestasiemetingsinstrumente gebruik kan word om die individuele ontwikkelingsbehoeftes van elke werknemer te bepaal. Die doel van die studie is die ontwikkeling en validering van 'n gedragsobservasieskaal vir senior bestuurders. Vir die doel is 49 senior bestuurders se prestasiemetings verkry op drie skale, naamlik 'n gedragsobservasieskaal wat ontwikkel is, die Werknemerbeoordelingskaal en 'n gedragsgeankerde skaal. 'n Pearson korrelasie van 0,74 is tussen die totaaltellings van die Gedragsobservasieskaal en die van die Gedragsgeankerde skaal verkry (p<0.01 ;N=49). Tussen die totaaltellings van die Gedragsobservasieskaal en die Werknemerbeoordelingskaal is 'n Pearson korrelasie van 0,75 gevind (p<0,01 ;N=49). Met hierdie resultate kan dus aanvaar word dat die Gedragsobservasieskaal 'n geldige meetinstrument van werksprestasie is en as prestasiemetingskaal gebruik kan word. / Research indicates that appraisal scales based on specific behaviour can be used as performance appraisal instruments to promote every employees' individual development need. The aim of this study is the development and validation of a behavioural observation scale for senior management. To achieve this aim 49 senior managers' performance scores were obtained on three scales, namely a behavioral observation scale which was developed, the Employee Rating Scale, and a behaviourally anchored rating scale. A Pearson correlation of 0,74 was obtained between the total scores of the Behavioral Observation Scale and the Behaviourally Anchored Rating Scale (p<0,01 ;N=49). Between the total scores of the Behavioral Observation and the Employee Rating Scales a Pearson correlation of 0,75 was found (p<0,01 ;N=49). With these results in mind, the Behavioral Observation Scale can be accepted as a valid measurement scale of work performance and can thus be applied as a performance appraisal instrument. / Industrial and Organisational Psychology / M. Comm. (Industrial Psychology)

Summaries in English and Afrikaans / Daar is gepoog om 'n omvattende, algemeen geldige maatstaf te vind waarmee enige opvoeding vir outentisiteit geevalueer kan word. Die moontlikheid dat die "essensiestruktuur van behoorlike volwassenewees en van opvoedingsdoel" (ES) hieraan voldoen, is ondersoek. Die genoemde essensiestruktuur (ES) is in 'n modeldiagram CESM) saamgevat. Dit bevat die volgende essensies: 'religiositeit', die 'etiese', 'affektiwiteit', 'liggaamlikheid'. 'nasionaliteit' (insluitende 'kulturaliteit'. 'historisiteit' en die 'politiese'), die 'estetiese'. 'individualiteit'. 'sosialiteit'. die 'ekonomiese'. 'vryheid'. 'gesag'. 'taal' en 'redelikheid'. asook die sewe normbeelde wat 'n substruktuur van die 'etiese' is. 'Religiositeit' is hierargies die mees belangrikste essensie. gevolg deur die 'etiese'. Aldie ander essensies verkeer op 'n derde niveau van belangrikheid. Om outentiek te wees behoort 'n opvoedingseksemplaar aan boge noemde hierargie te voldoen. Elk van die genoemde essensies is verduidelik, asook hulle samehangende verbandhoudendheid binne die ESM. Die harmonie binne hierdie gestruktureerdheid is van deurslaggewende belang vir die begryping van outentisiteit by opvoeding. 'n Evalueringsmetode. bekend as ''essensiestruktuur-analise" (ES-analise). is uit bogenoemde deur die navorser ontwikkel. 'n Uiteensetting van mikro-. meso- en makro ES-analise is oak verstrek. Beskrywings van eksemplare van opvoeding en van aanverwante opvoedingsaangeleenthede word in die lig van die ESM geanaliseer. Hoe meer 'n opvoedingseksemplaar ooreenstem met die ESM hoe nader beweeg dit aan outentisiteit. en omgekeerd. Ter illustrasie van die toepassingsmoontlikhede van ES-analise is die mensbeskouing van C.K. Oberholzer. die kindbeskouing van J.C. Coetzee. die Christelike lewensopvatting, die wereldbeskouing van Albert Schweitzer en die opvoedingsleer van die Klassisisme aan ES-analises onderwerp. Die Christelike lewensopvatting kon as outentiek geevalueer word, terwyl die mensbeskouing van Oberholzer. die kindbeskouing van Coetzee en die wereld beskouing van Schweitzer naby aan outentisiteit beweeg. Die opvoedingsleer van die Klassisisme vertoon heelparty essensie-diskrepansies. Die hipotese, dat ES-analise 'n betroubare evalueringsmaatstaf vir outentieke opvoeding kan wees, is bevestig. Derhalwe kon die aanwending vanESanalise oar 'n wye spektrum aanbeveel word. byvoorbeeld by histories opvoedkundige eksemplare, godsdienslere. skoolstelsels en meegaande kurrikula en by die ontwikkeling van opvoedingsprogramme. Die huidige opvoeding en kon temporere leefwyse kan in die toekoms verbeter word indien ES-analise daadwerklik toegepas sou word. / An effort was made to find a universally valid criterion to evaluate any education for authenticity. The "essence-structure of proper adulthood and of education-aim" (ES) was investigated in this regard. This essence-structure (ES) was condensed in a model diagram (ESM). It includes the fo 11 owing essences: · re 1 i gi ousness · , the · ethi ca 1 ' , · affec tiveness'. 'bodiliness'. 'nationality' ('culturality'. 'historicity' and the 'political'), the 'aesthetical'. 'individuality', 'sociality·. the 'economical'. 'freedom·. 'authority'. 'language' and 'rationality'. as well as the seven norm images - a substructure of the 'ethical'. Hierarchically 'religiousness· is the most important essence. followed by the 'ethical·. All other essences exist on a third 1 eve 1 of importance. To qualify as authentic an exemplar of education ought to comply with the above-mentioned hierarchy. Each of the essences and their interrelatedness within the ESM were discussed. The harmonious co-existence within this structuredness is of paramount importance in understanding educational authenticity. Emanating from the above-mentioned an evaluation method. known as "essence structure-analysis" (ES analysis). was developed. Micro. meso and macro-ESanalysis were also explained. Descriptions of exemplars of education and of educationally related matters are analysed using the ESM as the criterion. The more an exemplar corresponds with the ESM. the more it re sembles authenticity, and vice versa. To demonstrate the application potential of ES analysis C.K. Oberholzer's view of man. J.C. Coetzee's view on childhood. the Christian life-view. Albert Schweitzer's world-view and the educational doctrine of the Classicism were subjected toES analyses. The Christian life-view could be evaluated as authentic. while Oberholzer's view of man. Coetzee's view on childhood and Schweitzer's world-view are resembling authenticity closely. The educational doctrine of the Classicism revealed several essence discre pancies. (viii) The hypothesis. that ES analysis can be a reliable evaluation criterion for authentic education. was confirmed. It could thus be recommended that ESanalysis be implemented over a broad spectrum. for example when studying historico-educational exemplars. doctrines of religion. school systems and curricula and assisting in developing educational programs. The contemporary education and way of life can be ameliorated in the future if ES analysis is applied actively. / Educational Studies / D.Ed. (Philosophy of Education)

The advent of ARTs has enabled many individuals to have children and build families. Although ARTs have from the start been designated to serve as alternative way for heterosexual infertile individuals and couples to have genetically related children, ARTs are nowadays widely used by gays and lesbians to have even genetically unrelated children and build their families. This study addresses the well-being of children born as a result of ARTs and growing up in homosexual families in South Africa. South Africa has legalised homosexual unions, granting gays and lesbians several rights, including the right to marry, use ARTs to reproduce, and build families in which they raise their children. South Africa has also provided constitutional and statutory protection of children’s rights and has further required that the child’s best interests be considered as paramount in every matter concerning the child. Although ARTs may have allowed people to have children, they have proven to put the child’s interests at risk. ARTs are associated with several physical and psychological problems for resulting children. The legal protection provided for those children seems to be inadequate in respect of their best interests. Unlike Australian statutes that have provided strong protection for the child’s best interests, South African legislations regulating ARTs are far from protecting ART-born children’s interests. The application of the child’s best interests criterion to ART procedures has revealed that in the USA and Australia efforts of the state, ART providers and parents have been centred on the transfer of the custody of the ART-born child to the commissioning parent(s). Although in South Africa the application of the child’s best interests in the context of surrogacy procedures has revealed the protection of the child’s interests, it should be noted that that protection seems to focus on the child’s post-birth period. This situation leaves ART-born children without any protection, especially before their birth. In order to give effect to section 28 of the Constitution of the Republic of South Africa, 1996 and protect ART-born children’s interests, I make certain proposals for law reform in the final chapter of this thesis. / Private Law / LL. D

Le problème d'intérêt est d'estimer la fonction de concentration et l'aire sous la courbe (AUC) à travers l'estimation des paramètres d'un modèle de régression linéaire avec un processus d'erreur autocorrélé. On construit un estimateur linéaire sans biais simple de la courbe de concentration et de l'AUC. On montre que cet estimateur construit à partir d'un plan d'échantillonnage régulier approprié est asymptotiquement optimal dans le sens où il a exactement la même performance asymptotique que le meilleur estimateur linéaire sans biais (BLUE). De plus, on montre que le plan d'échantillonnage optimal est robuste par rapport à la misspecification de la fonction d'autocovariance suivant le critère du minimax. Lorsque des observations répétées sont disponibles, cet estimateur est consistant et a une distribution asymptotique normale. Les résultats obtenus sont généralisés au processus d'erreur de Hölder d'indice compris entre 0 et 2. Enfin, pour des tailles d'échantillonnage petites, un algorithme de recuit simulé est appliqué à un modèle pharmacocinétique avec des erreurs corrélées. / The problem of interest is to estimate the concentration curve and the area under the curve (AUC) by estimating the parameters of a linear regression model with autocorrelated error process. We construct a simple linear unbiased estimator of the concentration curve and the AUC. We show that this estimator constructed from a sampling design generated by an appropriate density is asymptotically optimal in the sense that it has exactly the same asymptotic performance as the best linear unbiased estimator (BLUE). Moreover, we prove that the optimal design is robust with respect to a misspecification of the autocovariance function according to a minimax criterion. When repeated observations are available, this estimator is consistent and has an asymptotic normal distribution. All those results are extended to the error process of Hölder with index including between 0 and 2. Finally, for small sample sizes, a simulated annealing algorithm is applied to a pharmacokinetic model with correlated errors.