pAIramid: Artificial intelligence for faster, smarter aerostructure certification

The aerospace sector faces increasing pressure to reduce emissions and accelerate the introduction of lighter, more efficient aircraft. However, developing and certifying new composite aerostructures remains time-consuming and costly, largely due to the extensive series of physical tests required to prove performance and safety. 

By creating a digital, AI-enhanced approach that interconnects different scales of material and structural testing – from the smallest composite sample to full-size aircraft components – pAIramid seeks to reduce the number of physical tests needed, accelerate design iterations and pave the way toward virtual certification in aviation.

pAIramid’s overall goal is to develop and validate a digital framework that integrates AI with multi-scale simulations and experimental data to enhance the traditional testing pyramid. The project will demonstrate how data-driven models can accelerate the design and verification of next-generation composite aerostructures, while maintaining reliability and compliance with certification standards.

Specific objectives include:

• Reducing product development time by up to 40% through AI-guided optimisation of the certification workflow;
• Enabling traceable data exchange and validation across testing levels, linking material, component and structural behaviour;
• Developing a federated learning framework that allows partners to share insights without disclosing sensitive data;
• Demonstrating the approach in four industrially relevant use cases covering both liquid resin infusion and fused deposition modelling composite technologies.

pAIramid methodology

The project’s concept builds on the traditional testing pyramid, which progresses from coupon level experiments to full-scale component validation. pAIramid enhances this framework by integrating AI-based virtual testing capabilities that connect experimental data and simulation models across all levels.

The pAIramid iTOOL will serve as a digital backbone to collect, manage and analyse data, allowing for predictive modelling and feedback between scales. It will combine physics-based models with machine learning to continuously refine predictions as new data becomes available. 

This approach ensures that knowledge gained from early tests informs higher levels of validation, creating a dynamic feedback loop that increases accuracy while reducing the physical testing requirements.

Four demonstrators will validate the system: a light-aircraft door and a wing leading edge manufactured using one-shot liquid resin infusion and a vertical stabiliser and wing leading edge produced by fused deposition modelling with continuous fibre reinforcement. Together, these use cases represent a wide range of design challenges and manufacturing processes, ensuring broad applicability of the results.

Expected impact

pAIramid is expected to have a strong impact on the competitiveness and sustainability of Europe’s aerospace industry. By integrating AI into the certification workflow, the project will significantly reduce the time and cost associated with developing new composite aerostructures. The ability to rely more heavily on validated virtual testing will also enable the introduction of novel materials and manufacturing routes, contributing to lighter, more efficient aircraft with lower emissions.

Beyond technological innovation, pAIramid supports a cultural shift towards data driven certification and closer collaboration between industry, regulators and research organisations. The project’s federated data approach ensures that sensitive industrial information remains protected while promoting knowledge sharing across the aviation ecosystem.

Ultimately, pAIramid will deliver a practical pathway toward virtual certification in Europe, aligning with the objectives of the Clean Aviation Joint Undertaking and advancing the transition to a greener, more agile aerospace sector.

For more on pAIramid, visit the project website.