The increasing functional complexity of composite structures requires material selection approaches that extend beyond traditional property-centric methods. The aim of this study is to establish and demonstrate a Structure–Property–Function (SPF) framework for function-oriented material selection, and to demonstrate its applicability through a composite blade case study using CNT-doped adhesive systems. The framework integrates representative volume element (RVE) modeling, microstructure-informed finite element simulations, and a Functional Area Clustering strategy to link material architecture directly to regional functional requirements. Mechanical and electrical properties were computed across 72 CNT–matrix combinations, revealing systematic trends in stiffness enhancement, percolation-driven conductivity, and sensitivity to CNT aspect ratio and volume fraction. Mapping these combinations onto multifunctional design spaces enabled the identification of viable material candidates for five critical blade regions, highlighting design trade-offs between stiffness, insulation, EMI shielding, and recyclability. By shifting the emphasis from material benchmarking to function-driven design, the proposed SPF framework offers a generalizable and scalable methodology for developing multifunctional composite components. The approach provides a foundation for future integration of optimization and experimental validation, supporting the development of next-generation composite structures in renewable energy and beyond. • SPF framework links CNT microstructure to functional adhesive performance. • RVE–FE modelling predicts stiffness and conductivity of CNT-doped adhesives. • Functional Area clustering identifies viable multifunctional material domains. • Trade-offs revealed among stiffness, insulation, EMI shielding and recyclability. • Function-driven selection enables targeted design of composite blade adhesives
Madia et al. (Sun,) studied this question.