Artificial Intelligence in Metal Additive Manufacturing: Applications in Design, Process Modeling, Monitoring, and...

Area of Science:

• Materials Science and Engineering
• Manufacturing Technology
• Artificial Intelligence

Background:

• Metal additive manufacturing (MAM) offers potential for complex component production in aerospace, energy, and biomedical fields.
• Widespread industrial adoption of MAM is hindered by defects, residual stresses, distortions, microstructural variability, and complex process parameters.

Purpose of the Study:

• To review the application of artificial intelligence (AI) across the metal additive manufacturing workflow.
• To analyze AI's role in addressing MAM challenges from design to microstructure prediction.
• To identify current advancements and persistent challenges in AI-driven MAM.

Main Methods:

• A structured literature review of studies from 2015-2025 was conducted using Scopus, Web of Science, and IEEE Xplore.
• Literature was analyzed across key MAM domains: design for additive manufacturing (DfAM), process modeling, in situ monitoring, and property prediction.
• AI approaches were categorized by learning paradigm (supervised, deep, reinforcement, hybrid physics-ML).

Main Results:

• AI, including machine learning and deep learning, is advancing parameter optimization, defect detection, and digital-twin frameworks for MAM.
• Recent AI applications show progress in areas like AI-assisted parameter optimization and process supervision.
• Key challenges remain, including data scarcity, heterogeneity, limited transferability, and the need for uncertainty-aware models.

Conclusions:

• The integration of multi-sensor monitoring with hybrid physics-informed AI models is the most promising path for near-term MAM improvements.
• This integration can enhance process reliability, reduce experimental iterations, and expedite industrial qualification.
• AI offers a pathway to overcome MAM limitations and accelerate its industrial adoption.