A knowledge based modelling system for the design and evaluation of flexible manufacturing facilities

The objective of this thesis is to explore the application of artificial intelligence (AI) techniques for modelling in flexible manufacturing. The work consists of three main parts. In the first part, the structure and performance of various types of flexibly automated batch manufacturing systems are discussed, the modelling challenge for the design of these types of manufacturing systems is identified, and the currently available modelling techniques are examined and comparatively assessed. In the second part, the research into the structure and design of a knowledge based modelling system is reported. Potential advantages of AI techniques for manufacturing systems modelling are identified. The modelling system is then developed using the LOOPS knowledge engineering language on the Xerox 1186 AI Workstation. Major features of the modelling system include its knowledge driven requirement to enab,l e evaluation of alternative systems with different criteria, the capability of modelling over multiple levels of detail, the transparency of its solution procedure, and the modularity of the system structure to allow convenient modification and extension. The third part is concerned with the evaluation of the AI based modelling method .. Parallel experiments are conducted on an extended case study cell by using the knowledge based modelling system, the emulator and the tool flow modelling system. Merits of the AI based method are then critically assessed, drawn on the comparison of the results obtained from the three studies. Conclusions drawn from this research and directions for future work are finally indicated.