No description available.

No description available.

No description available.

No description available.

No description available.

This thesis explores the architectural requirements of engineering design support systems based on knowledge based systems technology. The exploration is based on an understanding of the nature of designing as a professional activity and on the extent to which designers' competence can be modelled. Attention is focused on certain salient aspects of designers' competent behaviour. The theoretical study leads to the specification of requirements to be satisfied by a knowledge based system which will support designers in their professional setting and to the proposal of some knowledge based system components which will meet the requirements identified. The theoretical aspect of the thesis is complemented by a case study based on a designer of high voltage electricity distribution networks. The case study illustrates the theoretical component of the thesis and the methodological basis for the work. The practical realizability of the components of the knowledge based systems architecture proposed are demonstrated using the results of the analysis of the knowledge elicited in the case study without prejudicing the general applicability of the ideas. An object-oriented knowledge engineering software development environment is used to demonstrate how some components of the design situation represented can be implemented.

The application of computers in the manufacturing industry has substantially altered the control procedures used to program a whole manufacturing process. Currently, one the problems which automated manufacturing systems are experiencing is the lack of a good overall control system. The subject of this research has been centred on the identification of the problems involved in current methods of control and their advantages and disadvantages in an automated manufacturing system. As a result, a different type of control system has been proposed which distributes both the control and the decision making. This control model is an hybrid of hierarchical and hierarchical control systems which takes advantage of the best points offered by both types of control structures. The Durham FMS rig has been used as a testbed for an automated manufacturing system to which the hybrid control system has been applied. The implementation of this control system would not have been possible without the design and development of a System Integration Tool (SIT). The system is capable of real-time scheduling of the system activities. Activities within the system are monitored in real-time and a recording of the system events is available, which allows the user to analyse the activities of the system off-line. A network independent communication technique was developed for the Durham FMS which allowed the manufacturing cells to exercise peer-to-peer communication. The SIT also allowed the integration of equipment from different vendors in the FMS.

This dissertation addresses the problem of Computer Aided Architectural Evaluation and Design, with particular reference to building cost evaluation and cost modelling. It identifies the lack of building evaluation tools in the CAAD environment, and suggests an integrated approach to building modelling and cost modelling. The interaction of elemental and spatial descriptions of a design solution is considered as an original contribution to the field of computer aided building modelling and evaluation. It demonstrates the potential of CAAD and Bills of Quantities intergration to give an extra dimension to cost modelling at early design stages. Essentially, this research project advocates a larger overlap in the use of computers for the generation and evaluation of design. It asserts that any computer aided solution evaluation system must be able to converse with the designer during the highly integrated iteration of briefing, analysis, synthesis and evaluation stages of design. A working example is produced after software specification and implementation, to demonstrate to possibilities and/or limitations of such an approach.

The work reported in this thesis is related to the development of a generic model for a 'make-to-stock’ manufacturing organisation. This research is based upon the hypothesis that every type of manufacturing company should have a generic model. The objectives of the model are first, to help SNIEs (Small and Medium sized Enterprises) who cannot afford external help and second, to fulfil the need for modelling that arose from a review of modelling literature. These were the basic driving forces in carrying out this research and from which the objectives are derived. A survey of various system design and analysis methods was carried out. Initially SSADM, SADT, IDEFO, NMRISE, STRIM and GRAI were selected for the study. The reason for the selection of these methods was that each contained graphical tools and fulfilled the requirements for modelling. Further investigation showed GRAI and IDEFO to be the most suitable methods for modelling manufacturing systems and these were selected for further investigation. A direct comparison of the GRAI and IDEFO methods was carried out using a case study. The results of this indicated that the GRAI method was the most suitable for the analysis and design of manufacturing systems and demonstrated advantages over the IDEFO method. For this reason the GRAI method was selected and applied to case study 'make-to-stock! manufacturing organisations. The case studies demonstrated that several similar characteristics existed in 'make-to-stock’ manufacturing organisations, supporting the hypothesis of the research. The case studies also indicated that the application methodology had some significant drawbacks. On the basis of this investigation, a detailed methodology to apply the GRAI method was formulated and validated using a further detailed case study. After the methodology had been validated, it was used to develop a generic model for a 'make-to-stock' manufacturing organisation. The model developed presents the activities carried out at the strategic, tactical and operational levels of the management hierarchy. Details of typical decision horizon and review periods are also included in the model. The thesis presents the first detailed methodology for applying the GRAI method and first application to develop a generic model. It also presents the first detailed comparison between the IDEFO and GRAI methods. The findings of this research and recommendations for future work are presented in the final chapter.

No description available.