Session: DFMLC-03-03: Design for Additive Manufacturing

Paper Number: 143911

143911 - A Review of Traditional and AI-Driven Design for Additive Manufacturing 

Design for Additive Manufacturing (DfAM) is instrumental in harnessing the full spectrum of additive manufacturing (AM) by fine-tuning part design to exploit its unique capabilities while navigating its inherent limitations. This paper delves into a thorough review of DfAM, bridging both traditional methodologies and contemporary machine learning (ML) paradigms. By embedding manufacturing insights at the inception of design, DfAM facilitates the optimization of parts tailored for AM processes, addressing critical considerations such as material choice, structural integrity, and the imperative for support structures. This review sets itself apart by meticulously dissecting rule-based design principles, tailored DfAM processes, and the role of ML in augmenting DfAM. It shines a spotlight on ML's capacity to forecast manufacturing defects, refine design parameters, and spur innovation through generative design techniques. Moreover, the paper tackles the intricacies and challenges endemic to DfAM, like the convoluted nature of the solution landscape and the scarcity of robust data sets, advocating for a future lined with research into sophisticated ML models and cross-disciplinary collaborations. This contribution is pivotal to the discourse in DfAM, offering a panoramic view of its methodologies, the transformative impact of ML, and charting out the trajectory for future explorations and innovations within the DfAM ecosystem.

Presenting Author: Binyang Song Virginia Tech

Presenting Author Biography: Dr. Binyang Song is an Assistant Professor in the Industrial and Systems Engineering at Virginia Tech. With a Ph.D. in Data-Driven Design from Singapore University of Technology and Design, Dr. Song conducted research on Human-AI Hybrid Teaming and Multi-modal Learning for engineering design at the Penn State and MIT during her postdocs. Her research features an interdisciplinary approach, emphasizing collaboration between human and technology in complex problem solving.

Authors:

Binyang Song Virginia Tech