The air transportation system is composed of multiple elements and belongs to a complex socio-technical system. It is difficult to assess the risk of an aircraft fault because it could constantly change during operation and is influenced by numerous factors. Although traditional methods such as Failure Mode, Effects, and Criticality Analysis (FMECA) and Fault Tree Analysis (FTA) can reflect the degree of fault risk to a certain extent, they cannot accurately quantify and evaluate the fault risk under the multiple influences of human factors, random faults, and external environment. In order to solve these problems, this article proposes a fault risk assessment method for aircraft control systems based on a fault risk composite assessment framework using the Improved Risk Priority Number (IRPN) as the basis for the fault risk assessment. Firstly, a Bayesian network (BN) and Gated Recurrent Unit (GRU) are introduced into the traditional evaluation framework, and a hybrid prediction model combining static and dynamic failure probability is constructed. Subsequently, this paper uses the functional resonance analysis method (FRAM) by introducing a risk damping coefficient to analyze the propagation and evolution of fault risks and accurately evaluate the coupling effects between different functional modules in the system. Finally, taking the fault of a jammed flap/slat drive mechanism as an example, the risk of the fault is evaluated by calculating the IRPN. The calculation results show that the comprehensive failure probability of the aircraft control system in this case is 3.503 × 10⁻⁴. Taking into account the severity, the detection, and the risk damping coefficient, the calculation result of IRPN is 158.00. According to the classification standard of the risk level, the failure risk level of the aircraft belongs to a controlled risk, and emergency measures need to be taken, which is consistent with the actual disposal decision in this case. Therefore, the evaluation framework proposed in this article not only supports a quantitative assessment of system safety and provides a new method for fault risk assessments in aviation safety management but also provides a theoretical basis and practical guidance for optimizing fault response strategies.

To explore the coupling among the safety, economic, and environmental objectives in flight operations control, the trade-off patterns among these objectives were analyzed. The core scope was defined, key factors and coupling feedback relationships were identified, and a system dynamics (SD) model with seven stocks was constructed, based on which a semi-quantitative trade-off analysis was conducted. A fully connected neural network was then introduced to fit the nonlinear mapping between six policy levers and four core objectives, and a multi-objective optimization model was built using the elitist Non-dominated Sorting Genetic Algorithm II (NSGA-II), generating the Pareto front. The constraint relationships were quantified via dynamic trend analysis, threshold attainability tests, and compromise and trade-off coefficients, and the stability of the conclusions was verified by robustness tests. The mean compromise coefficient was 0.957, and the mean profit–safety, carbon emissions–profit, and carbon emissions–safety trade-off coefficients were 12.335, 4.757, and −4.110, respectively, indicating trade-offs among the safety, economic, and environmental objectives. The study finds that under a given total resource budget, policy space, and the airline’s existing operating mode, the three objectives cannot simultaneously attain their individual optima within the feasible solution set. A strong trade-off exists between safety and economy, while the impacts of environmental performance on the other two objectives are constrained by policy settings and technological pathways, offering theoretical references for airlines’ operational decision-making.