Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experiments

Achieving desired mechanical properties in additive manufacturing requires many experiments and a well-defined design framework becomes crucial in reducing trials and conserving resources. Here, we propose a methodology embracing the synergy between high-throughput (HT) experimentation and hierarchi...

Full description

Saved in:
Bibliographic Details
Main Authors: Mahsa Amiri, Zahra Zanjani Foumani, Penghui Cao, Lorenzo Valdevit, Ramin Bostanabad
Format: Article
Language:English
Published: Elsevier 2025-04-01
Series:Materials & Design
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S026412752500125X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849339146682761216
author Mahsa Amiri
Zahra Zanjani Foumani
Penghui Cao
Lorenzo Valdevit
Ramin Bostanabad
author_facet Mahsa Amiri
Zahra Zanjani Foumani
Penghui Cao
Lorenzo Valdevit
Ramin Bostanabad
author_sort Mahsa Amiri
collection DOAJ
description Achieving desired mechanical properties in additive manufacturing requires many experiments and a well-defined design framework becomes crucial in reducing trials and conserving resources. Here, we propose a methodology embracing the synergy between high-throughput (HT) experimentation and hierarchical machine learning (ML) to unveil the complex relationships between a large set of process parameters in Laser Powder Bed Fusion (LPBF) and selected mechanical properties (tensile strength and ductility). The HT method envisions the fabrication of small samples for rapid automated hardness and porosity characterization, and a smaller set of tensile specimens for more labor-intensive direct measurement of yield strength and ductility. The ML approach is based on a sequential application of Gaussian processes (GPs) where the correlations between process parameters and hardness/porosity are first learnt and subsequently adopted by the GPs that relate strength and ductility to process parameters. Finally, an optimization scheme is devised that leverages these GPs to identify the process parameters that maximize combinations of strength and ductility. By founding the learning on larger “easy-to-collect” and smaller “labor-intensive” data, we reduce the reliance on expensive characterization and enable exploration of a large processing space. Our approach is material-agnostic and herein we demonstrate its application on 17-4PH stainless steel.
format Article
id doaj-art-56e432a4b3db4d09a0f935807a6724c3
institution Kabale University
issn 0264-1275
language English
publishDate 2025-04-01
publisher Elsevier
record_format Article
series Materials & Design
spelling doaj-art-56e432a4b3db4d09a0f935807a6724c32025-08-20T03:44:12ZengElsevierMaterials & Design0264-12752025-04-0125211370510.1016/j.matdes.2025.113705Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experimentsMahsa Amiri0Zahra Zanjani Foumani1Penghui Cao2Lorenzo Valdevit3Ramin Bostanabad4Materials and Manufacturing Technology Program, University of California, Irvine, CA 92697, USADepartment of Mechanical and Aerospace Engineering, University of California, Irvine, CA 92697, USAMaterials and Manufacturing Technology Program, University of California, Irvine, CA 92697, USA; Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA 92697, USA; Department of Materials Science and Engineering, University of California, Irvine, CA 92697, USAMaterials and Manufacturing Technology Program, University of California, Irvine, CA 92697, USA; Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA 92697, USA; Department of Materials Science and Engineering, University of California, Irvine, CA 92697, USA; Corresponding author at: Department of Materials Science and Engineering, University of California, Irvine, CA 92697, USA.Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA 92697, USA; Department of Civil and Environmental Engineering, University of California, Irvine, CA 92697, USA; Corresponding author at: Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA 92697, USA.Achieving desired mechanical properties in additive manufacturing requires many experiments and a well-defined design framework becomes crucial in reducing trials and conserving resources. Here, we propose a methodology embracing the synergy between high-throughput (HT) experimentation and hierarchical machine learning (ML) to unveil the complex relationships between a large set of process parameters in Laser Powder Bed Fusion (LPBF) and selected mechanical properties (tensile strength and ductility). The HT method envisions the fabrication of small samples for rapid automated hardness and porosity characterization, and a smaller set of tensile specimens for more labor-intensive direct measurement of yield strength and ductility. The ML approach is based on a sequential application of Gaussian processes (GPs) where the correlations between process parameters and hardness/porosity are first learnt and subsequently adopted by the GPs that relate strength and ductility to process parameters. Finally, an optimization scheme is devised that leverages these GPs to identify the process parameters that maximize combinations of strength and ductility. By founding the learning on larger “easy-to-collect” and smaller “labor-intensive” data, we reduce the reliance on expensive characterization and enable exploration of a large processing space. Our approach is material-agnostic and herein we demonstrate its application on 17-4PH stainless steel.http://www.sciencedirect.com/science/article/pii/S026412752500125XLaser powder bed fusionHigh-throughput experimentsGaussian processUncertainty quantification17-4PH stainless steelMechanical properties
spellingShingle Mahsa Amiri
Zahra Zanjani Foumani
Penghui Cao
Lorenzo Valdevit
Ramin Bostanabad
Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experiments
Materials & Design
Laser powder bed fusion
High-throughput experiments
Gaussian process
Uncertainty quantification
17-4PH stainless steel
Mechanical properties
title Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experiments
title_full Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experiments
title_fullStr Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experiments
title_full_unstemmed Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experiments
title_short Unveiling processing–property relationships in laser powder bed fusion: The synergy of machine learning and high-throughput experiments
title_sort unveiling processing property relationships in laser powder bed fusion the synergy of machine learning and high throughput experiments
topic Laser powder bed fusion
High-throughput experiments
Gaussian process
Uncertainty quantification
17-4PH stainless steel
Mechanical properties
url http://www.sciencedirect.com/science/article/pii/S026412752500125X
work_keys_str_mv AT mahsaamiri unveilingprocessingpropertyrelationshipsinlaserpowderbedfusionthesynergyofmachinelearningandhighthroughputexperiments
AT zahrazanjanifoumani unveilingprocessingpropertyrelationshipsinlaserpowderbedfusionthesynergyofmachinelearningandhighthroughputexperiments
AT penghuicao unveilingprocessingpropertyrelationshipsinlaserpowderbedfusionthesynergyofmachinelearningandhighthroughputexperiments
AT lorenzovaldevit unveilingprocessingpropertyrelationshipsinlaserpowderbedfusionthesynergyofmachinelearningandhighthroughputexperiments
AT raminbostanabad unveilingprocessingpropertyrelationshipsinlaserpowderbedfusionthesynergyofmachinelearningandhighthroughputexperiments