부산대학교 도서관

Due to the complexity of the actual product structure, parting design has been reliant on the experience of designers excessively which becomes a gigantic obstacle to automation design of mold. In practice, the determination of parting direction usually depends on the assessment of different parting structure. In this paper, lots of information properties such as uncertainty, inconsistency, and preference relationship existing in the evaluation criteria are extracted by a variety of means. A ranking mechanism based on preference rough set model is proposed to reduce the redundant data and get the comprehensive dominance of each scheme. Finally, a case is used to illustrate the feasibility of the method, and the optimal parting result is suitable for the practice. [ABSTRACT FROM AUTHOR]