Toward data-driven research: preliminary study to predict surface roughness in material extrusion using previously published data with Machine Learning

TL;DR

This work addresses the challenge of predicting surface roughness in material extrusion without large experimental campaigns by leveraging multi-source published data. It implements a data-driven ML pipeline, training on literature-derived data and validating against new experiments, with Random Forest delivering the best performance (correlation 0.93 on literature data and 0.79 on new data; MAPEs of 13% and 8%, respectively). The study identifies layer height as the most influential factor and demonstrates that published data can support accurate surface roughness modeling across printers and environments. The results highlight the potential of data-driven approaches to reduce time and cost in additive manufacturing while outlining limitations and avenues for future research, including larger datasets, additional process factors, and deep learning techniques for improved generalization.

Abstract

Material extrusion is one of the most commonly used approaches within the additive manufacturing processes available. Despite its popularity and related technical advancements, process reliability and quality assurance remain only partially solved. In particular, the surface roughness caused by this process is a key concern. To solve this constraint, experimental plans have been exploited to optimize surface roughness in recent years. However, the latter empirical trial and error process is extremely time- and resource-consuming. Thus, this study aims to avoid using large experimental programs to optimize surface roughness in material extrusion. Methodology. This research provides an in-depth analysis of the effect of several printing parameters: layer height, printing temperature, printing speed and wall thickness. The proposed data-driven predictive modeling approach takes advantage of Machine Learning models to automatically predict surface roughness based on the data gathered from the literature and the experimental data generated for testing. Findings. Using 10-fold cross-validation of data gathered from the literature, the proposed Machine Learning solution attains a 0.93 correlation with a mean absolute percentage error of 13 %. When testing with our own data, the correlation diminishes to 0.79 and the mean absolute percentage error reduces to 8 %. Thus, the solution for predicting surface roughness in extrusion-based printing offers competitive results regarding the variability of the analyzed factors. Originality. As available manufacturing data continue to increase on a daily basis, the ability to learn from these large volumes of data is critical in future manufacturing and science. Specifically, the power of Machine Learning helps model surface roughness with limited experimental tests.

Figures (6)

• Figure 1: System scheme (figure by authors).
• Figure 2: Surface roughness profilometer (figure by authors).
• Figure 3: Measuring position and location of the zipper (top view of the printing bed) (figure by authors).
• Figure 4: Histogram of the prediction error for both training and testing (figure by authors).
• Figure 5: Surface roughness versus the layer height and wall thickness for both the Ultimaker 5S and Ultimaker 3 extended (figure by authors).