The rapid design of simulation models using cladistics and template based modelling

摘要

The drive towards a more globally competitive market has led to an increase in demandfor goods and services on a global scale. As a result of this increase in productivity,production systems are being designed and redesigned at an increased rate and that theyare becoming more innovative as they progress with time.The challenge this research attempts to address is how to improve the ability of the UKbasedmanufacturing industry to make a more effective decision during manufacturingsystems design/redesign by adopting simulation techniques as both strategic andoperational decision making tools. The aim of this research is to develop a classificationscheme based on cladistics and evolutionary analysis and to use this classification in thedevelopment of a template based modelling library. The research focused on identifyingthe existing manufacturing layout types, the various layout configurations that are beingused and template based model generation. Some of the major developments of theresearch conducted were the construction of the manufacturing layout and componentbased cladograms and the RapidSim generator.A literature review on manufacturing systems layouts revealed the types of systemlayouts which are most commonly used in the manufacturing sector as well as thecomponent configurations and characteristics which are found within each productionsystems. This research makes a major contribution by providing a cladisticalclassification of manufacturing systems layouts, an external interface for model buildingand development and a set of recommendations, which when adopted may help increasethe use of template based simulation modelling. Based on the data analysis carried out,the findings suggest that there is room for the development and implementation of atemplate based modelling approach to the development of simulation based models. Themost important result obtained from the validation of the model was that the time takento build and run the model decreased significantly by around 65% when compared to theconventional model building process.