About this Abstract |
| Meeting |
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
| Symposium
|
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
| Presentation Title |
Data-Driven Geometry Classification for Manufacturing Process Planning |
| Author(s) |
Matthew Youseff, Jeremy LeMaster, Jason Orellana-borja, Saurabh Basu, Sepideh Abolghasem |
| On-Site Speaker (Planned) |
Matthew Youseff |
| Abstract Scope |
This study presents a data-driven approach for classifying 3D part geometries with respect to their intended manufacturing processes. Our approach is based on Principal Component Analysis (PCA) of descriptors that were derived from these geometries by finding their intersections with a simulated growing sphere and fitting those intersections to spherical harmonics. A dataset of 160 parts comprising 120, and 40 3D geometries that were originally designed to be manufactured by turning, and milling, respectively, was compiled and analyzed. Further, the approach was tested on the raw harmonics besides statistics extracted from them, including maximum, mean, standard deviation, and entropy. Initial results demonstrated that PCA-transformed spherical harmonic features can effectively cluster 3D geometries with respect to their intended manufacturing processes for which they were designed. The geometric basis of this effect was discussed along with how the results obtained from this approach may be utilized to infer the cost of manufacturing. |
| Proceedings Inclusion? |
Planned: Post-meeting proceedings |