Quantitative suggestions for build orientation selection

Abstract

Orientation determination is an essential process planning task in additive manufacturing (AM) since it directly affects the part quality, part texture, mechanical properties, build time, fabrication cost, etc. Evaluation method provides a simple and effective way to determine the optimum orientation of a part. However, the candidates are predesigned in the evaluation method, which results in limited candidate space and makes the qualities of evaluation results highly dependent on the qualities of the predesigned candidates; bad outcomes will be obtained due to the poor candidates. To this end, a feedback multi-attribute decision-making (MADM) model is proposed in this work. The feedback MADM model is an integration of two sub-models: MADM model and proportional-integral-derivative (PID) control model. MADM model aims to calculate the score of a given build orientation. Three criteria, surface roughness, support volume, and build time, are considered in this model, and the ordered weighted averaging (OWA) operator is applied for aggregation. In the PID control model, the finite candidate space is first expanded to infinity by quaternion rotation, then PID controller is applied to match the build orientation with the user’s requirements, that is, search for the part orientation whose score is consistent with the user expected score in the infinite alternative orientations. Four parts with different geometric structures are tested in the experiments, and evaluation and control features of the feedback MADM model are discussed.