| Abstract Scope |
Since the articulation of process-based quality in welding and additive manufacturing in 2000, advances have moved the field toward qualification grounded in science rather than empirical trial-and-error. This raises a fundamental question: Are we there yet? Drawing on recent work in in-situ sensing, diffraction, residual stress prediction, microstructure evolution, process consistency metrics, and data-driven monitoring, this keynote examines how physics-based models are increasingly coupled with high-dimensional data and artificial intelligence. Across fusion and solid-state processes, these advances reveal both the promise and the limits of current qualification pathways. AI is accelerating anomaly detection, similarity assessment, and process control; however, trustworthy qualification still depends on mechanistic understanding of heat transfer, solidification, phase transformation, distortion, and structure–property relationships. The central argument is that AI will not replace process-based qualification, but strengthen it into an adaptive, evidence-rich framework that enables the democratization of qualification intelligence. |