Machine learning guided design of experiments to accelerate exploration of a material extrusion process parameter space

Devin Young, Britannia Vondrasek and Michael W. Czabaj ()
Additional contact information
Devin Young: University of Utah
Britannia Vondrasek: University of Utah
Michael W. Czabaj: University of Utah

Journal of Intelligent Manufacturing, 2025, vol. 36, issue 1, No 28, 508 pages

Abstract: Abstract Parts produced using material extrusion (MEX), a common additive manufacturing method, are often limited to non-structural applications due to sub-optimal mechanical properties, including poor interlayer fracture toughness, Gc. Gc of MEX parts depends on process parameters, but the complex relationships between process parameters and Gc are not well understood. This paper describes the use of a machine learning (ML) method using Forests with Uncertainty Estimates for Learning Sequentially (FUELS) to study the effect of five process parameters on the Gc of MEX parts. Training data for the FUELS model is collected using a modified double cantilever beam (MDCB) test, and Gc is calculated using a classical beam theory approach. The FUELS method provides guided testing by suggesting additional parameter combinations from high-uncertainty regions of the parameter space. After sequentially testing a total of 2.9% of the 2205 possible parameter combinations, there was minimal change in the non-dimensional model error, and training was concluded. Gc values collected from testing ranged 0.056 kJ/m2 to 1.774 kJ/m2. The resulting parameter space was examined to better understand how Gc evolves with changing process parameters. Among other results, extrusion temperature was shown to have a greater effect on Gc at higher print speeds. Overall, the FUELS method, paired with accelerated experimental testing, provides a useful means of quickly exploring the large MEX parameter space to establish relationships among process parameters and Gc. The methods of this study can serve as a blueprint for other studies with large parameter spaces, not just in MEX but in other manufacturing processes.

Keywords: Additive manufacturing; Machine learning; Interlayer fracture toughness; ABS; FUELS (search for similar items in EconPapers)
Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10845-023-02255-5 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:joinma:v:36:y:2025:i:1:d:10.1007_s10845-023-02255-5

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10845

DOI: 10.1007/s10845-023-02255-5

Access Statistics for this article

Journal of Intelligent Manufacturing is currently edited by Andrew Kusiak

More articles in Journal of Intelligent Manufacturing from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().