Parting Direction Selection and Evaluation Using an Extension of Fuzzy MCDM Method

Parting direction selection and evaluation plays an important role in injection mold design. In order to select suitable parting direction, several criteria need to be taken into account including undercut feature, parting line complexity, draw depth, project area, and flatness. This paper proposes an extension of fuzzy multi-criteria decision making (MCDM) approach to support parting direction selection and evaluation process. In proposed approach, the rating of alternatives and important weight of criteria are expressed linguistic terms. We then develop the membership functions for the final fuzzy evaluation values of alternatives by the interval arithmetic of fuzzy numbers. Next, this paper uses the new left and right indices ranking approach to defuzzify the normalized weighted ratings into crisp values to determine the ranking order of alternatives. Finally, a numerical example is used to illustrate the applicability and efficiency of the proposed method.

Determination of Parting Lines and Multiple-Undercut Boundaries for Complex CAD Parts with Application to Multi-Piece Molds Design

This paper proposes a new slicing approach that uses multiple sets of cutting planes to automatic determine parting lines and boundaries of undercuts of complex CAD parts for the multi-piece molds design. In this algorithm, the group of feasible parting directions is first collected and three sets of cutting planes are then formed based on the primary parting direction. One set of cutting planes is used to generate slicing profiles, and two others are used to extract intersection points passed through by parting loops and boundaries of undercuts. The result shows that the proposed parting lines are suitable to obtain parting surfaces and construct the entire mold-pieces. The algorithm provides a highly effective solution for handling many complex industrial parts.

NON-DISCRETIZED APPROACH TO VISIBILITY ANALYSIS FOR AUTOMATIC MOLD FEATURE RECOGNITION USING STEP PART MODEL

Undercut feature recognition in automatic mold design requires analyzing the visibility/accessibility of the part along a known parting direction, the existing approaches for which are mostly restricted to polyhedral parts or a particular CAD platform or cannot handle partially accessible faces very well. The present work proposes a projection-based methodology to analyze the visibility of a part from a given parting direction without discretizing the part. An approach to further extract undercut features and generate corresponding side-cores is also presented. The implemented system uses STEP AP214/AP203 part files for input and output and has been successfully validated using parts geometrically similar to those used in literature as well as on a number of industrial parts of greater complexity.