@ARTICLE{Kumar_A._Comparison_2025,
 author={Kumar, A. and Das, D. and Raj, N.D. and Bag, S. and Srivastava, V.C.},
 volume={vol. 70},
 number={No 1},
 pages={29-38},
 journal={Archives of Metallurgy and Materials},
 howpublished={online},
 year={2025},
 publisher={Institute of Metallurgy and Materials Science of Polish Academy of Sciences},
 publisher={Committee of Materials Engineering and Metallurgy of Polish Academy of Sciences},
 abstract={Wire arc additive manufacturing (WAAM) is amongst the emerging technologies of the layer-by-layer deposition process to manufacture the metallic parts. Multi-layer bead deposition using the WAAM process leads to the fabrication of complex products with practical utility. The bead profile of each layer controls the geometry of the final product. However, the melting efficiency and the bead geometry depend on the various process parameters. The primary process parameters affecting the melting efficiency and the bead geometry are the wire feed rate, travel speed, diameter of the wire, and power. Owing to the various complexities during metal deposition, predicting the dimensions at each deposition attribute is not always feasible. Hence, the current work is focused on the utilization of different machine learning (ML) algorithms to understand the relationship between the process parameters, melting efficiency, and bead geometry. The different ML models used for the current work are linear regression (LR), decision tree regressor (DTR), random forest (RF), support vector regression (SVR) and, extra tree regressor (ETR). The ETR is found to predict the melting efficiency with the highest prediction rate of 97.4%, whereas, the SVR and LR predict the bead width and height with the highest accuracy rate of 97.4% and 98.7%, respectively.},
 title={Comparison of Machine Learning Performance for the Prediction of Melting Efficiency and Bead Geometry in wire Arc Additive Manufacturing Process},
 type={Article},
 URL={http://ochroma.man.poznan.pl/Content/134444/AMM-2025-1-04-Kumar.pdf},
 doi={10.24425/amm.2025.152503},
 keywords={WAAM, bead geometry, melting efficiency, machine learning regression models, hyperparameter tuning},
}