Akademik Veri Yönetim Sistemi
II. INTERNATIONAL CONFERENCE ON SCIENTIFIC AND INNOVATION RESEARCH, Sivas, Türkiye, 15 - 17 Eylül 2023, ss.1226-1241
The aim of this study is to present a model for estimating part production times in threedimensional (3D) printer using data-based machine learning approach. Production time is
a critical factor in ensuring that 3D printers are used effectively and efficiently. In this
study, analysis of different parameters (material properties, geometry complexity, printer
settings, etc.) affecting production time was made. Machine learning algorithms were used
along with the analysis of existing datasets and the training process. The data collection
process includes the information obtained during the production of experimental parts on
3D printers with various features. These data are correlated with actual production times
as well as important parameters such as printer settings, material properties, printing
geometry. The model is designed in such a way that it can process large-scale and complex
data sets and provide high accuracy in predictions. Machine learning algorithms include
widely used methods such as artificial neural networks and support vector machines. These
algorithms are trained to estimate production times by analyzing the properties and
relationships of datasets. In addition, in this study, the analysis and importance of the
parameters affecting the production time are emphasized. In addition to material properties,
the effects of factors such as geometry complexity, support strategy, layer thickness on the
production time were investigated. This model provides 3D printer users with a valuable
tool to predict and optimize print times.