Hybrid modelling methodology for system design
In the face of rapid development in information technology coupled with a growing dynamism in global markets, manufacturing systems have to be re-constructed for short term or long term goal. Such innovations promise to lead to a new competitive stage, which typically involve design of function, information and behaviour of systems. In order to design the system, simulation has often been chosen. However, simulation has proved limited and fails to aid design of such a complex systems because of consuming much computing time and cost, especially when modelling larger systems. Thus, there is a need to seek a new approach, in a way that results in simulating such a large manufacturing system with less demand on computing time and cost. This study researches into a hybrid modelling approach to minimise these limitations. It includes proposing a hybrid modelling methodology and developing a hybrid modelling tool. The methodology integrates simulation and metamodelling techniques. The metamodel employed in the study possesses, not only characteristics of conventional metamodels in terms of representing relationships in quantity, but also in time lapse. This is the originality of the study and the significant distinction between this research and application of metamodelling in conventional ways. The hybrid modelling tool is developed to support and demonstrate the identified hybrid methodology. LISP has been used as the software language for the hybrid modelling tool. The result of this work concludes that the hybrid modelling approach is capable of simulating a complex manufacturing system with less demands on the computer. The work reported in this thesis has been carried out in conjunction with the EPSRC research project, Hierarchical Manufacturing System Modelling (HMSM) (GR/F96549), to produce an Integrated Design and Modelling Methodology (IDEM). The project was initially a collaborative research program including Loughborough University of Technology (LUT), Morris Crane Ltd., of Loughborough and GEC Large Machine, of Rugby. The experience of these collaborators has proved most valuable in supporting the research, and have provided a cross section of views and comments. The research reported in this thesis is set in the context of the HMSM Research group at Loughborough.