Evaluation of the Additive Manufacturability of CAD-Parts for initial Data Labelling in AI-based Part Identification

Abstract

The objective of this study is to develop and implement an analysis tool that identifies CAD geometries which impair Additive Manufacturing. A key performance indicator is generated to be used as data label representing the manufacturability for a future application of AI. Relevant geometric features are identified and algorithms to evaluate critical features are developed. The analyses include part orientation, build volume, wall thicknesses, gap widths, bore and cylinder diameters as well as the process-specific factors powder removal and need for support structures. The manufacturability of a part is calculated as Additive Manufacturing Feasibility Indicator (AMFI), depending on the identified critical features and a user-specific weighting. The AMFI successfully serves as data label which is suitable for application in AI methods. The developed GUI supports designers by highlighting critical features directly in the CAD environment and allowing the user in a purposeful part optimization for AM.

Evaluation of design support tools for additive manufacturing and conceptualisation of an integrated knowledge management framework

“…Additionally, identified critical features can be visualised. [53] Tendencies to make use of a KB as observed for mesh-based analyses can be seen here as well, but to a much lesser extent. Layer or slice-based analysis methods probably have the same problems as described for the universal approaches (see 2.1.2.1).…”

A knowledge-driven, integrated design support tool for additive manufacturing

“…Existing related approaches with CAD integration (Han and Schaefer, 2019;Winkler et al, 2021a;Winkler et al, 2021b) are either rather abstract or have practical limitations. (Ellsel et al, 2021) Two newer publications describe frameworks for knowledge-driven DfAM.…”