Efficient Categorization of Binder Jetting Printer Settings Using ANN Sensitivity Evaluation

Abstract

Abstract This study employs artificial neural network (ANN) sensitivity analysis to rank the impact of key binder jetting parameters, namely AFS grain fineness, printhead speed, drop mass, and print resolution (DX), on the strength of 3D-printed sand moulds. Results indicate that AFS grain fineness accounts for more than 70% of the influence on mould strength, with the remaining parameters contributing 30%. Leveraging these findings, an efficient categorization was developed. By ranking parameters through cumulative scoring, this classification highlights the relative importance of each variable. The resulting classification offers foundries a strategic tool to optimize binder jetting processes, adaptable to different machines and parameters. This approach advances innovation in the foundry industry, aligns with Fourth Industrial Revolution (4IR) technologies, and supports the United Nations Sustainable Development Goal 9, promoting industry, innovation, and infrastructure development.