Lightweight Structural Design of UAM Fuselage Using AI Predictive Modeling and Composite Big Data from Automated...

Traffic congestion and air pollution caused by rapid urbanization have emerged as critical challenges in metropolitan areas worldwide. Urban air mobility (UAM), particularly electric propulsion-based systems, has gained attention as a promising solution. For the successful commercialization of UAM, a lightweight airframe design with ensured structural integrity is essential. This study proposes an optimized lightweight design process that integrates automated composite manufacturing with artificial intelligence (AI)-based material property prediction. Finite-element analysis (FEA) was performed on glass fiber-, basalt fiber-, and carbon fiber-reinforced polymers under identical deformation conditions to derive design material properties in terms of elastic modulus and weight reduction. A large-scale dataset of fiber-reinforced plastics was established through an automated manufacturing process, and a deep learning regression model was developed using Altair AI Studio to predict mechanical properties under untested material and process conditions. The predicted properties were applied to a UAM fuselage model, and FEA results demonstrated that composite structures achieved equivalent or superior stiffness with up to 50% weight reduction compared to aluminum. In addition, inverse reserve factor (IRF) analysis confirmed structural safety, with all configurations maintaining IRF values below 1. The proposed AI-driven framework provides a scalable and data-driven lightweight design methodology applicable to next-generation UAM and advanced air mobility structures.