Research On Reconfiguration Of Multi-Station Machine Tool Using Case-based Reasoning

Reconfigurable Manufacturing System (RMS) offers a cost-effective, highly productive method of assuring productivity growth and competitive responsiveness. Combined with the project research on reconfigurable design theory and technology on multi-station numeric control and diagnosis devices (No.502056) which is sponsored by the Zhejiang Province Natural Science Fund, the design methodology for RMS based on the multi-station machine tool (MSMT) is investigated in this dissertation using Case-based Reasoning (CBR).Chapter 1, the state-of-the-art of RMS research is discussed and analyzed. Proposed in 1996, RMS is new agile manufacturing system in order to quickly respond to market changes. Research on RMS is categorized into system-level and machine-level in which the Engineering Research Center at University of Michigan plays a leading role.Chapter 2, the buildup and design process of MSMT are expounded. The design process of MSMT, including process planning, layout design, station machine tool selection and design, material handling device design, control system design, etc., is characterized by experience-based which is consistent with CBR. Hence, the case-based method is proposed for MSMT reconfiguration.Chapter 3, MSMT reconfiguration process model is established which contains part process characteristics and process plans comparative studies, layout reconfiguration, station machine tool adjustment, material handling device reconfiguration, etc. Case-based process is composed of case retrieval, case reuse, case revising and case retaining.Chapter 4, case representation for MSMT is studied. MSMT case is composed of part, MSMT and reconfiguration strategy organized in hierarchical and linear structure. Feature-based, frame and rule-based method are utilized in representation for part, MSMT and reconfiguration strategy respectively.Chapter 5, case retrieval and reuse for MSMT are studied. An updated algorism for fuzzy feature similarity computation is proposed. Then, an acquirement algorithm for difference feature is given in order to carry on the case comparative studies.Chapter 6, case revising for MSMT is investigated. The case revising model aimed at solving difference feature, based on reconfiguration strategy (adaptation case) is set up. A new method using analytical hierarchy process (AHP) is proposed to support station machine tool selection and its analytical model is built.Chapter 7 summarizes the work of the dissertation. The shortcoming and future work are also sketched out.