Page 25 - IJAMD-2-2

International Journal of AI for
Materials and Design AI-driven material development for AM

Figure 12. An overview of AI-driven material development for AM
Abbreviations: AM: Additive manufacturing; AI: Artificial intelligence.

predictive modeling, data-driven material discovery, and bioprinting, AI plays a critical role in optimizing bioink
process optimization, significantly improving efficiency formulations by balancing printability, biocompatibility,
and material performance. and mechanical integrity. It has also been applied to adjust
This review highlights the transformative role of AI in extrusion parameters through reinforcement learning,
material development for AM, with particular emphasis on thereby improving reproducibility and promoting cell
material design and performance optimization. Traditional viability in printed constructs. Across these domains, AI
trial-and-error approaches are inefficient and costly, efforts are directed toward intrinsic material development,
whereas AI – particularly ML and DL – enables predictive with a focus on selecting compositions, controlling phase
modeling of material behavior, composition optimization, evolution, and enhancing functional performance. This
and microstructural tailoring. The integration of AI with emphasis distinguishes material-centric strategies from
physics-based methods, such as DFT, CALPHAD, and broader manufacturing optimization. Moreover, AI
FEA, further enhances the accuracy and efficiency of improves the understanding of complex PSP relationships,
material development workflows. which enables the design of materials with reliable and
tailored properties for AM applications.
The success of AI-driven approaches depends critically
on the availability of high-quality datasets. These datasets In conclusion, AI offers a paradigm shift in AM-oriented
can be obtained through experimental measurements, material development by accelerating the design of
physics-based simulations, and structured online materials with tailored compositions and microstructures,
repositories. However, most existing databases have been aligned with the specific requirements of AM processes.
developed for conventional manufacturing and do not Future progress depends on improved data availability,
capture the process-specific features required for AM. closer integration between AI and physical modeling,
This limitation is particularly evident under conditions and a sustained focus on the core material properties that
involving rapid solidification and non-equilibrium phase define performance.
transformations. To address this gap, there is a pressing
need to construct AM-oriented datasets and apply rigorous 7. Concluding remarks and perspectives
data pre-processing procedures. As discussed above, the continued advancement of AI
AI applications in AM material development span metals, is expected to revolutionize material development in
polymers, and bioprinting. In metal AM, AI facilitates AM by significantly enhancing predictive capabilities.
alloy design, phase prediction, and microstructure control, By integrating AI with AM, researchers can accelerate
which helps improve printability and enhance mechanical material discovery, optimize processing conditions, and
properties. In polymer AM, AI-guided generative design improve overall performance. Looking ahead, several key
supports the development of mechanical metamaterials advancements are poised to shape the future of AI-driven
with improved structural performance. In the context of AM materials development, as presented below.

Volume 2 Issue 2 (2025) 19 doi: 10.36922/IJAMD025100007