Methodical development of data driven quantitative design rules in additive manufacturing

Abstract

Additive manufacturing (AM) and its industrial application are growing steadily.However, despite many research efforts, there are challenges in the practical implementation.In addition to methodological challenges within the design process, the detailed design of AM components, which is suited for manufacturing as well as economically beneficial, poses difficulties.As an improvement of the situation, a methodical framework is presented, through which design rules for a practice-oriented design of AM geometries can be developed.The design rules developed based on this framework are of qualitative as well as quantitative nature.The qualitative perspective is concerned with the question of whether the geometries can be manufactured depending on their features.The quantitative perspective is concerned with the quality of the manufactured geometries depending on their features.Within this framework, the potential influence of process parameters on part quality is addressed.The method is based on applied statistics on the one hand and uses methods of explorative data analysis on the other hand.A case study is conducted to demonstrate the framework.This study consists of the development of a quantitative design rule for a geometry element (hollow cylinder) with reference to a material extrusion process.The results of the case study are statistical models for various quality characteristics (diameter deviations and cylindricity).The terms of the models are obtained by regression and are then statistically tested for significance.The models set up also allow an assessment of the achievable tolerance grades for geometric features investigated and their respective quality characteristics.