ETD Collection
Permanent URI for this collectionhttps://wiredspace.wits.ac.za/handle/10539/104
Please note: Digitised content is made available at the best possible quality range, taking into consideration file size and the condition of the original item. These restrictions may sometimes affect the quality of the final published item. For queries regarding content of ETD collection please contact IR specialists by email : IR specialists or Tel : 011 717 4652 / 1954
Follow the link below for important information about Electronic Theses and Dissertations (ETD)
Library Guide about ETD
Browse
6 results
Search Results
Item A complexity management approach for designing viable IT service systems in South Africa(2019) Mokgala, Sekhwela MosesThis research investigates important determinants in designing, implementing and managing viable Information Technology (IT) service systems in the South African economy where digitalisation and globalisation are introducing unprecedented market and technological dynamics. The research adopts the Viable Systems Approach to provide a systematic, iterative process to service system design, tackling both service delivery and service co-creation complexities through the application of Viable System Theory. Having identified and discussed various existing service system design models available in the literature to date, the residual complexities these models present are discussed, and the research proposes an integrated model for designing viable IT service systems. The developed Integrated Service System Model is used for developing a complexity management approach to designing viable IT service systems within selected firms in South Africa. The finding is that the Viable Systems Approach efficiently reveals residual variety in operational IT services systems under investigation while also providing viable service system designs for services systems found to contain a residual variety. The use of the Cybernetics model within the Viable Systems Approach proves useful in analysing specified service system viability operating in dynamic markets by providing the analytical tools and methods to interrogate systemic behaviour from a complexity perspective. The recommendation is that in erratic dynamic markets where previous performance history does not guarantee sustainable performance, residual variety presented by the market environment and the ability to effectively implement viable service system designs are critical determinants to designing viable IT service systems that can survive in the digital economy. The implications for management is the foundational understanding and a multi-disciplined application of Organisational Cybernetics (OC) in the design and management of service systems. Further research is suggested in applying the developed Integrated Service System Model in other industries and service system context to prove or improve the tenacity of the model in designing service system that adapts to dynamic market environments over the entire service system lifecycle.Item An expert system for selecting roofing types in developing areas(1993) Oliveira, Silvia Da Conceicao CunhaThis dissertation has as it's objective to develop an expert system for the purpose of investigating the suitability of using expert systems in developing areas. It is the first time to our knowledge that an expert system has been tested for usage in such a context. The housing and more specifically roofing domain is investigated for an application topic resulting in "the selection of suitable roofing types" being chosen. Potential end-users are identified, an appropriate context of such a system is outlined and an expert system tool is selected, The knowledge engineering and building of the knowledge base are described. Feedback from the parties involved in testing and validating BUILDROOF is documented. In addition the suitability of using expert systems in the developing areas context in which BUILDROOF was developed, is discussed. Finally, recommendations regarding potentially suitable areas of application for expert systems in developing areas are outlined.Item A real-time expert system shell for process control.(1990) Kang, Alan MontzyA multi-layered expert system shell that specifically addresses real-time issues is designed and implemented. The architecture of this expert system shell supports the concepts of parallelism, concurrent computation and competitive reasoning in that it allows several alternatives to be explored simultaneously. An inference engine driven by a hybrid of forward and backward chanining methods is used to achieve real-time response, and certainty factors are used for uncertainty management. Real-time responsiveness is improved by allowing the coexistence of procedural and declarative knowledge within the same system. A test bed that was set up in order to investigate the performance of the implemented shell is described. It was found in the performance analysis that the proposed system meets the real-time requirements as specified in this research.Item Influence modelling and learning between dynamic bayesian networks using score-based structure learning(2018) Ajoodha, RiteshAlthough partially observable stochastic processes are ubiquitous in many fields of science, little work has been devoted to discovering and analysing the means by which several such processes may interact to influence each other. In this thesis we extend probabilistic structure learning between random variables to the context of temporal models which represent partially observable stochastic processes. Learning an influence structure and distribution between processes can be useful for density estimation and knowledge discovery. A common approach to structure learning, in observable data, is score-based structure learning, where we search for the most suitable structure by using a scoring metric to value structural configurations relative to the data. Most popular structure scores are variations on the likelihood score which calculates the probability of the data given a potential structure. In observable data, the decomposability of the likelihood score, which is the ability to represent the score as a sum of family scores, allows for efficient learning procedures and significant computational saving. However, in incomplete data (either by latent variables or missing samples), the likelihood score is not decomposable and we have to perform inference to evaluate it. This forces us to use non-linear optimisation techniques to optimise the likelihood function. Furthermore, local changes to the network can affect other parts of the network, which makes learning with incomplete data all the more difficult. We define two general types of influence scenarios: direct influence and delayed influence which can be used to define influence around richly structured spaces; consisting of multiple processes that are interrelated in various ways. We will see that although it is possible to capture both types of influence in a single complex model by using a setting of the parameters, complex representations run into fragmentation issues. This is handled by extending the language of dynamic Bayesian networks to allow us to construct single compact models that capture the properties of a system’s dynamics, and produce influence distributions dynamically. The novelty and intuition of our approach is to learn the optimal influence structure in layers. We firstly learn a set of independent temporal models, and thereafter, optimise a structure score over possible structural configurations between these temporal models. Since the search for the optimal structure is done using complete data we can take advantage of efficient learning procedures from the structure learning literature. We provide the following contributions: we (a) introduce the notion of influence between temporal models; (b) extend traditional structure scores for random variables to structure scores for temporal models; (c) provide a complete algorithm to recover the influence structure between temporal models; (d) provide a notion of structural assembles to relate temporal models for types of influence; and finally, (e) provide empirical evidence for the effectiveness of our method with respect to generative ground-truth distributions. The presented results emphasise the trade-off between likelihood of an influence structure to the ground-truth and the computational complexity to express it. Depending on the availability of samples we might choose different learning methods to express influence relations between processes. On one hand, when given too few samples, we may choose to learn a sparse structure using tree-based structure learning or even using no influence structure at all. On the other hand, when given an abundant number of samples, we can use penalty-based procedures that achieve rich meaningful representations using local search techniques. Once we consider high-level representations of dynamic influence between temporal models, we open the door to very rich and expressive representations which emphasise the importance of knowledge discovery and density estimation in the temporal setting.Item A framework for a real-time knowledge based system.(1993) Gebbie, IanA framework designed to contain and manage the use of knowledge in a real-time knowledge based system for high level control of an industrial process is presented. A prototype of the framework is designed and implemented on a static objectorientated shell. Knowledge is stored in objects and in forward chaining rules. The knowledge has a well defined structure, making it easy to create and manage. Rules are used to recognize conditions and propose control objectives. The framework uses the knowledge to determine variables that if altered will meet the objectives. Control actions are then found to implement changes to these variables The use of explicit control objectives makes it possible to determine if an action worked as intended and if its use is suitable for the present conditions. This enables a learning mechanism to be applied in the expert system. The prototype operated adequately, but the knowledge required to drive the. system was found to be very detailed and awkward to create.Item An expert system approach to decision modelling for savanna management(2016-07-18) Berliner, Derek DavidNo abstract provided.