Computer-Aided Intelligent Product Design Process and Practice

Abstract

Modern product design leverages computational software and intelligent systems to create high-performance and cost-effective products. In aerospace, materials must be continually made lighter and resistant to ensure structural integrity and safety. Traditional material selection processes rely heavily on expert knowledge and do not balance multiple performance criteria. This work develops an artificial intelligence-supported computer-aided model to forecast the accommodation of materials in aerospace structures and recommend the best solutions for the lightweight and high-strength parts. The approach to designing an aircraft landing gear strut assembly system involves using material data from the designed airship and applying an XGBoost classification model with Pareto optimisation, which ranks and predicts the most suitable materials for the aircraft. The best-performing materials are tested using finite element analysis using vibration/modal testing, and thermal testing to ensure structural integrity under operational conditions. Findings show the model achieved 98.1% accuracy in predicting aerospace-suitable materials based on key mechanical properties. Also, the selected material (Aluminium 7075-T6) exhibits a thermal conductivity of 130 W/m·K. These results confirm the framework’s effectiveness in enabling data-driven, cost-efficient, and reliable material selection, thereby streamlining the structural design process and advancing intelligent product design practices in aerospace engineering.