Dagal, Idriss2026-01-312026-01-3120260951-83201879-0836https://doi.org./10.1016/j.ress.2025.111841https://hdl.handle.net/20.500.12662/10667The increasing complexity of next-generation avionic systems necessitates advanced reliability frameworks to ensure fault tolerance while meeting stringent constraints on weight, cost, and certification. This paper presents a novel probabilistic modeling approach that integrates Dynamic Fault Tree Analysis (DFTA), Bayesian Belief Networks, and semi-Markov processes to assess failure probabilities in safety-critical flight control architectures, such as Fly-by-Wire (FBW) systems and Integrated Modular Avionics (IMA). To complement this, a Dynamic Redundancy Optimization (DRO) framework using a Multi-Objective Genetic Algorithm (MOGA) is introduced to optimally allocate redundancy under realistic conditions, including common-cause failures and intermittent faults. The methodology is validated on a triplex-redundant Flight Control Computer (FCC), achieving a 41.8 % improvement in mean time between failures (MTBF) and a 36.5 % reduction in catastrophic event probability compared to static baseline models. Importantly, the framework conforms to DO-178C and ARP4761A standards, ensuring traceability and certification readiness. The results reveal a Pareto frontier representing optimal tradeoffs between reliability enhancements and system resource overheads, providing critical guidance for the design of next-generation avionic systems and regulatory assessment.eninfo:eu-repo/semantics/closedAccessDynamic fault tree analysis (DFTA)Bayesian networks (BNs)Redundancy optimizationFlight control systemsAvionics reliabilityArticle10.1016/j.ress.2025.1118412-s2.0-105020942559Q1266WOS:001608386000009Q1