Machine Learning Model Selection for Performance Prediction in 3D Printing

Prediction of elongation, roughness, tensile strength yield of a 3D printed product is important to boost the quality of the printing process. This paper inspects and compares most machine learning regression strategies to build a robust representation that best models the 3D printing process. 3D printing process is affected by nine primary input factors such as layer height, wall thickness, infill density, infill pattern, nozzle temperature, bed temperature, print speed, material and fan speed. These factors influence the quality of the 3D printed product. Multiple machine learning models are run based on the experiments and the best model is selected based on the prediction performance. Among the various machine learning strategies, the best is identified which best predicts the output factors. It was inferred that “radial basis function regressor” algorithm was able to best predict the results for elongation with a mean absolute error of 0.3421 and root mean square error of 0.3421. “Additive regression” was able to best predict the results for elongation with a mean absolute error of and 28.2969 and root mean square error of 39.2895. “Random committee regression” was able to best predict the results for tensile strength with a mean absolute error of 4.112 and root mean square error of 5.2789. © 2022, The Institution of Engineers (India).