Certification-Informed Reliability and Performance Assessment of Electrified Aircraft

Regulatory standards for emerging electric and hybrid-electric aircraft require demonstrating safe flight and landing under component failures or degraded performance. Meeting these requirements early in the design process remains challenging as traditional conceptual design workflows often treat reliability as a secondary, post hoc consideration. This paper introduces a computational methodology that integrates component-level reliability, system-level failure analysis, and certification-driven performance evaluation for a given aircraft configuration. A Bayesian approach is used to propagate uncertain component-level failure rates through the powertrain architecture and quantify the likelihood of critical failure states, which are then assessed against the reliability limits specified in 14-CFR. For failure scenarios with exceedance probabilities above allowable thresholds, a 6-DoF flight dynamics model is employed to compute the degraded-performance response and compare it directly to the required regulatory performance margins. The method is demonstrated using the NASA X-57, a distributed propulsion concept with an electrified powertrain. Results show that wingtip motor failures occur with probabilities exceeding the allowable 1�� −7 per flight hour requirement, and that under these failures the aircraft fails to achieve the required 8.3% climb gradient, as specified in §23.2120(a)(1), indicating non-compliance with Part 23. The approach is formulated to be fully compatible with gradient-based MDAO through the use of computational graphs, enabling scalability to larger design spaces and complex architectures. This work provides a pathway for certification-informed conceptual design and sets the stage for future geometry and architecture optimization to achieve regulatory compliance.