Analysis and evaluation of some manufacturing processes and systems using graph theory and matrix approach

Abstract

Manufacturing is defined as the transformation of material and information into goods for the satisfaction of human needs, and is one of the primary wealth generating activities for any nation and contributes significantly to the employment. The study of the issues related to manufacturing becomes increasingly important not only to the industry but also to the academic community, which is often challenged with finding solutions to the problems relevant to the industry. In order to optimize the systems/processes, various analyses are required. These analyses of manufacturing system can be conceptualized as the mapping from the performance requirements of manufacturing system onto suitable values of decision variables, which describe the physical design of system. This thesis provides an integrated approach called graph theory matrix approach for the evaluation of manufacturing system or process. This thesis further explores the capability of digraph and matrix approach to analyze and evaluate the manufacturing systems/processes. This approach provides an index, which shows the performance of the systems/processes. A software has been developed to find the universal evaluation index from the universal attributes characteristic permanent function. Two case studies which have been considered in this research work for the justification of employing the graph theory and matrix approach in the manufacturing process/system applications. They are the evaluation of jig for flange coupling and the evaluation and selection of the best design for a motor bike kicker from the defined two alternatives based on design aspects. In this research work, the milling machine selection problem has also been considered in which the AHP, ANP and graph theory and matrix approach have been employed and subsequently the comparison of the results implies that graph theory and matrix approach has more effectiveness than that of the AHP and ANP.