Multidisciplinary Surrogate-Based Structural Optimization of a Full Aircraft Configuration

Designing lightweight and structurally efficient aircraft remains a critical challenge in aerospace engineering, particularly for composite aircraft configurations. Towards this effort, this study presents a surrogate-assisted structural sizing framework for full-aircraft configurations, integrating Finite Element Modeling (FEM), structural analysis and multi-disciplinary constraints to optimize the thickness distribution of various structural components. Unlike conventional approaches that rely on simplified structural models or optimization of isolated components, the proposed framework relies on a computationally efficient structural optimization process while employing high-fidelity modeling and accurately capturing global load interactions across the entire aircraft level. The framework optimizes the thickness of primary structural components, including wing skins, spars, fuselage skins, frames and tail, to minimize weight while satisfying static, buckling, and aeroelastic constraints. To reduce the computational cost associated with high-fidelity optimization, a surrogate modeling approach is adopted, enabling efficient exploration of the design space and accelerated convergence. Two strategies for sampling and surrogate model updates are examined: one using a large initial sample set with fewer infill points, and another with a smaller initial sample set but a greater number of infill points. Results indicate that both strategies successfully achieve feasible lightweight configurations, but with distinct trade-offs in accuracy and computational efficiency. The large-sample strategy converges faster but tends to over-constrain the design, while the adaptive infill strategy produces lighter and more structurally balanced solutions at the cost of increased evaluations. For both plans, buckling stability appears as the dominant constraint, thus driving the optimal thickness distribution, with composite failure and aeroelastic requirements frequently active but less restrictive. The analysis further highlights how structural stiffness requirements govern fuselage and wingbox reinforcement, while strength constraints primarily influence local skin and spar sizing. The predicted displacement levels, failure indices, and flutter speeds of the optimized configurations are consistent with trends reported for CRM/uCRM-based studies in the literature, providing qualitative validation of the structural and aeroelastic response. These findings demonstrate the capability of the proposed framework to capture critical structural interactions, guide design trade-offs, and enable efficient multidisciplinary optimization of full-aircraft configurations.