Dynamic fuzzy expert system for multi objective criteria for selection of manufacturing method using mamdani model

Installing a manufacturing method might be very expensive and time consuming project. Organizations should examine and decide on how best to make this decision of selecting appropriate process meeting their requirements. In order to improve the manufacturing cycle more than 110 manufacturing processes have been proposed. The objectives aimed at and the functions focused on by these processes vary. The process should be flexible enough to accommodate reasonable changes in design. This poses a great challenge to a manager in selection of effective and economical manufacturing process. Different organizations have different objectives and based on their specific requirement they deploy suitable process conforming to their objective. Today's business scenario is highly competitive, complex and dynamic in nature which demands strategic planning meeting the challenges of changing time. Recently, the authors have developed a tool to enable the end user a quick selection of appropriate manufacturing method based on multiple objectives. The end user instead of querying the database directly will use the natural language, termed as Manufacturing Query Language (MQL) designed by the authors, which is interfaced with RDBMS using prolog. The methodology adopted is based on crisp sets which does not take into account uncertainties in model parameters and changing business scenarios. To incorporate ambiguity into the system, the authors have developed a fuzzy expert system dynamically based on the differing ranges of triangular membership functions. In this paper, the authors present an architecture for dynamic fuzzy expert system for multi objective criteria for selection of manufacturing method using Mamdani Model. Fuzzy expert system is created outside MATLAB and MATLAB is used only for creating user interface for querying methods based on objectives and for the evaluation of rules. A simulink model is developed for selective methods and objectives and is executed for different combinations of objectives in class S. The input is derived from a excel file which is converted into a MAT file as required by MATLAB. Similary, the output stored in MAT file is converted into an .xlsx file. The large number of xlsx. Files required are generated using Excel Object Library which is interfaced with Java. The results obtained using Fuzzy Expert System are compared with that obtained using crisp expert system. Mamdani type FIS offers more flexibility in method selection due to the nature of output members functions which are overlapping fuzzy sets. This yields the manager a greater freedom in method selection based on infrastructure and other resource availability.