The industrial adoption of Digital Twins (DTs) for optimizing Additive Manufacturing (AM) processes is frequently hindered by the lack of standardized, reproducible frameworks. This report addresses this gap by proposing a structured methodology for the design and deployment of an end-to-end Digital Twin (DT) architecture for Laser Powder Bed Fusion (LPBF) processes. The core of this methodology is the synergistic fusion of two components: a physics-based simulation model based on Finite Element Model (FEM) that captures the complex physics phenomena of the process, and an AI-powered model for in-situ visual defect detection during the printing process. The entire system is orchestrated by a microservices backbone utilizing containerization and a modular big-data architecture that is realized via an event-driven message broker that ensures seamless data communication and interoperability. The aim of this work is to provide a clear, actionable guide and a generalized replication roadmap as a blueprint for researchers and engineers to systematically develop and deploy sophisticated DT systems, thereby bridging the gap between theoretical concepts and industrial viability.
Meintanis et al. (2025) studied this question.