What does this research mean for the field?

While machine learning in directed energy deposition (DED) shows high reported accuracies, primarily using supervised learning on experimental data, its real-world deployment is limited by challenges such as scarce public datasets, weak generalization, and limited real-time implementations. Novelty: ClaimNovelty.SYNTHESIS. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

Synthesize the state of machine learning applications across directed energy deposition processes and identify key challenges limiting industrial deployment.

August 19, 2026Journal of Manufacturing ProcessesOpen Access

Machine learning in directed energy deposition: A systematic literature review

View Full Paper

Ask AI

Bookmark

Authors

HFHooman Fallahi MBMohammad Hassan Baqershahi HMHessamoddin Moshayedi

Discussion

Member takes

Overview

Systematic review reveals high predictive accuracy alongside deployment bottlenecks in machine learning for directed energy deposition, highlighting the need for standardized benchmarks.

Key Points

Synthesize the state of machine learning applications across directed energy deposition processes and identify key challenges limiting industrial deployment.
Systematic literature review synthesizing N=148 peer-reviewed studies published between 2016 and 2025.
Categorized literature by application domain (defect metrics, geometry prediction, mechanical properties, melt-pool characterization, thermal-field modeling) and learning methodology.
Supervised learning dominates the field, while reinforcement learning remains rare, multimodal fusion represents a growing minority, and approximately 75% of studies rely on experimental data.
Models frequently report high predictive accuracy (R² > 0.95 or classification accuracy > 90%), but practical implementation is limited by scarce public datasets, high annotation costs, and weak cross-machine generalization.

Cite This Study

Fallahi et al. (2026) studied this question.

synapsesocial.com/papers/6a8562eb03308d306e2d5e56 https://doi.org/10.1016/j.jmapro.2026.07.081

View Full Paper

Ask AI

Bookmark

August 19, 2026Journal of Manufacturing ProcessesOpen Access

Machine learning in directed energy deposition: A systematic literature review

View Full Paper

Ask AI

Bookmark

Authors

HFHooman Fallahi MBMohammad Hassan Baqershahi HMHessamoddin Moshayedi

Discussion

Member takes

Overview

Systematic review reveals high predictive accuracy alongside deployment bottlenecks in machine learning for directed energy deposition, highlighting the need for standardized benchmarks.

Key Points

Synthesize the state of machine learning applications across directed energy deposition processes and identify key challenges limiting industrial deployment.
Systematic literature review synthesizing N=148 peer-reviewed studies published between 2016 and 2025.
Categorized literature by application domain (defect metrics, geometry prediction, mechanical properties, melt-pool characterization, thermal-field modeling) and learning methodology.
Supervised learning dominates the field, while reinforcement learning remains rare, multimodal fusion represents a growing minority, and approximately 75% of studies rely on experimental data.
Models frequently report high predictive accuracy (R² > 0.95 or classification accuracy > 90%), but practical implementation is limited by scarce public datasets, high annotation costs, and weak cross-machine generalization.

Cite This Study

Fallahi et al. (2026) studied this question.

synapsesocial.com/papers/6a8562eb03308d306e2d5e56 https://doi.org/10.1016/j.jmapro.2026.07.081

View Full Paper

Ask AI

Bookmark