Modeling, Analysis, Evaluation and Selection of an Agile Manufacturing System

Abstract

Agile manufacturing systems are rapidly increasing in number due to ever changing customer demands, shorter product life cycles, higher product variety, economy of production and highly competitive global market. Global competition has forced manufacturing companies to decrease response time (for changes in design, operation, failures etc.) and also reduce lead time to the market. Also, the product life cycle in case of many products has shortened considerably due to continuous advancement in manufacturing technology and variety of products present in the market. So, a manufacturer requires high response, lower lead times along with economical production and best quality. The principles of agile manufacturing fulfill these requirements very well with characteristics such as re-configurability, reusability and scalability. In this thesis work, an integrated systems model for the structure of the agile manufacturing system in terms of its constituents and interactions between the constituents is developed using graph theoretic approach. The different constituents and the interaction between the constituents are identified and have been represented by graph-based model. The information of the graph is converted into matrix form. The matrix models and the variable permanent function models are developed for carrying out decomposition, characterization, and the total analysis. The terms of permanent multinomial characterize the agile manufacturing system uniquely and are highly useful for computational storage, retrieval, communication, as well as analysis of the structural information of agile manufacturing system. To illustrate the effectiveness of the Graph Theoretical Approach, an illustrative example has also been presented. A new methodology has been developed for the evaluation, comparison, ranking and optimum selection of an agile manufacturing system from the different alternative designs available in the global market. This proposed methodology is based on Multi Attribute Decision Making (MADM) approach. Pertinent attributes which describe the whole agile manufacturing system are identified in an exhaustive way. A n-digit coding scheme based on all the pertinent attributes for a given agile manufacturing system is suggested, that is useful for the development of a comprehensive database of available agile manufacturing systems, and their subsequent retrieval. The agile manufacturing system selection procedure allows rapid convergence from a very large number of options to a manageable iii shortlist of potentially suitable agile manufacturing systems using, elimination search based in few critical selection attributes. Then the selection procedure proceeds to evaluate and rank the shortlisted alternatives by employing the TOPSIS approach. The proposed 3-stage selection methodology is explained with the help of an illustrative example.