Towards Learning-Aided Configuration in 3D Printing

Abstract

Configurators rely on logical constraints over parameters to aid users and determine the validity of a configuration. However, for some domains, capturing such configuration knowledge is hard, if not infeasible. This is the case in the 3D printing industry, where parametric 3D object models contain the list of parameters and their value domains, but no explicit constraints. This calls for a complementary approach that learns what configurations are valid based on previous experiences. In this paper, we report on preliminary experiments showing the capability of state-of-the-art classification algorithms to assist the configuration process. While machine learning holds its promises when it comes to evaluation scores, an in-depth analysis reveals the opportunity to combine the classifiers with constraint solvers.

Effect of Feature Subset Selection on Samplings for Performance Prediction of Configurable Systems

“…Ridge Regression [Rajan 2022, Chang et al 2017, and others such as Random Forest [Amand et al 2019, Bao et al 2018], which is central to our methodology for feature selection. Our goal is to efficiently select essential features for machine learning models to balance resource usage and accuracy.…”

Input sensitivity on the performance of configurable systems an empirical study

“…We also publish the resulting datasets online (see the links in the Data column) and in the companion repository with replication details. 6…”

VariMod

“…The work by Amand et al [2] introduces a learning-based approach for detecting invalid parameter values of 3D models constructed by OpenSCAD scripts, and for deriving constraints on these parameters. The derived constraints improve the variability of 3D models by preventing errors resulting from incompatible parameter values.…”