Data-driven hierarchical decision support for civil aviation maintenance safety risk: A fusion of Bayesian network and system dynamics

  • Jirong Duan orcid

    School of Safety Science and Engineering, Civil Aviation University of China, Tianjin 300300, China
  • Ming Cheng orcid

    School of Safety Science and Engineering, Civil Aviation University of China, Tianjin 300300, China
Article ID: 3774
DOI: https://doi.org/10.59400/adecp3774

(This article belongs to the Special Issue Mathematical Analysis Advances in System Fault Analysis, Prediction and Control (Close))

Keywords: continuous endurance risk management; Bayesian network; system dynamics; risk analysis; data-driven

Abstract

To enhance the scientificity and precision of risk analysis and management decision-making in aircraft maintenance operations, this study proposes a risk analysis-decision model tailored to maintenance events. Based on actual civil aviation maintenance scenarios, the model employs real data to conduct data-driven analysis and precisely calculates the occurrence probabilities of various risk factors by constructing a Bayesian risk probability network. Meanwhile, it selects three categories of key risk factors: personnel (A), management (B), and organization (C), to build a system dynamics scenario, thereby simulating the long-term implementation effects of different management strategies. The research findings indicate that the existing maintenance management system demonstrates a certain level of risk buffering efficacy under normal operating conditions, effectively preventing risks from evolving into higher severity levels. The combinations of key risk factors at different severity levels exhibit a hierarchical characteristic, specifically manifesting as three tiers dominated by organization and safety barriers, personnel capabilities and behaviors, and daily operations and slow-variable risks, respectively. It is proposed that maintenance safety risk governance should adopt a graded and differentiated management strategy. At the decision-making level, the model is capable of simulating the long-term impacts of different management strategies. The study reveals that increasing management investment can significantly reduce process risks, whereas systemic risks and frontline operational errors require sustained, long-term resource allocation for improvement.

Published
2026-02-05
How to Cite
Duan, J., & Cheng, M. (2026). Data-driven hierarchical decision support for civil aviation maintenance safety risk: A fusion of Bayesian network and system dynamics. Advances in Differential Equations and Control Processes, 33(1). https://doi.org/10.59400/adecp3774
Issue
Vol. 33 No. 1 (2026)
Section
Articles

Copyright (c) 2026 Jirong Duan, Ming Cheng

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

References

[1]Strengthened profitability expected in 2025 even as supply chain issues persist. Available online: https://www.iata.org/en/pressroom/2024-releases/2024-12-10-01/ (accessed on 16 November 2025).

[2]Esteban A, Zafra A, Ventura S. Data mining in predictive maintenance systems: A taxonomy and systematic review. WIREs Data Mining and Knowledge Discovery. 2022; 12(5): e1471.

[3]Stanton I, Munir K, Ikram A, et al. Predictive maintenance analytics and implementation for aircraft: Challenges and opportunities. Systems Engineering. 2023; 26(2): 216–37.

[4]Bauranov A, Rakas J. Bayesian network model of aviation safety: Impact of new communication technologies on mid-air collisions. Reliability Engineering & System Safety. 2024; 243: 109905.

[5]Wang L, Wang Y, Ding Y. Risk analysis and resilience of humanitarian aviation supply chains: A Bayesian network approach. Applied Sciences. 2025;15(19): 10508.

[6]Pan W, Li Y, Jiang Y, et al. Analysis of controller-caused aviation accidents based on association rule algorithm and Bayesian network. Applied Sciences. 2025; 15(17): 9690.

[7]Zhang X, Mahadevan S. Bayesian network modeling of accident investigation reports for aviation safety assessment. Reliability Engineering & System Safety. 2021; 209: 107371.

[8]Zhou Z, Yu X, Zhu Z, et al. Development and application of a Bayesian network-based model for systematically reducing safety risks in the commercial air transportation system. Safety Science. 2023; 157: 105942.

[9]Guo R, Wu J, Ji F, et al. Analysis of traffic safety in airport aircraft activity areas based on Bayesian networks and fault trees. Digital Transportation and Safety. 2024; 3(1): 8–18.

[10]Liu H, Hu M, Yang L. A new risk level identification model for aviation safety. Engineering Applications of Artificial Intelligence. 2024; 136(Part A): 108901.

[11]Dang X, Shao Y, Liu H, et al. Risk Analysis Method of Aviation Critical System Based on Bayesian Networks and Empirical Information Fusion. Electronics. 2025; 14(12): 2496.

[12]Chen YH, Lin JJ. Determinants of flight delays at East Asian airports from an airport, route and network perspective. Journal of Air Transport Management. 2021; 94: 102064.

[13]Barry DJ. Estimating runway veer-off risk using a Bayesian network with flight data. Transportation Research Part C: Emerging Technologies. 2021; 128: 103180.

[14]Qi X, Liu C, Fan D, et al. Causal modeling and risk assessment of runway excursion events via interpretable machine learning. Reliability Engineering & System Safety. 2025; 269: 112055.

[15]Zhao Y, Hua S, Ren X. Relevance research of threat/error and undesired states in air traffic management based on Bayesian network model. Journal of Air Transport Management. 2017; 60: 45–48.

[16]Arnaldo Valdés RM, Gómez Comendador VF, Perez Sanz L, et al. Prediction of aircraft safety incidents using Bayesian inference and hierarchical structures. Safety Science. 2018; 104: 216–230.

[17]Ifelebuegu AO, Awotu-Ukiri EO, Theophilus SC, et al. The application of Bayesian—Layer of Protection Analysis method for risk assessment of critical subsea gas compression systems. Process Safety and Environmental Protection. 2018; 113: 305–18.

[18]He S, Shi S, Lin Z, et al. Risk assessment of coal mine gas explosion based on cloud model and Bayesian network. Reliability Engineering & System Safety. 2025: 111798.

[19]Wang N, Yuen KF, Li D, et al. A bayesian network approach to ship safety assessment: Integrating machine learning and expert opinions. Reliability Engineering & System Safety. 2025; 111659.

[20]Zhao L, Gao Q, Dong Y, et al. Vision-guided dynamic risk assessment for long-span PC continuous rigid-frame bridge construction through DEMATEL–ISM–DBN modelling. Buildings. 2025; 15(24): 4543.

[21]Li X, Wang G, Gui Y. Enhancing subway fire safety with a symmetric framework: From fault tree analysis to dynamic Bayesian network inference. Symmetry. 2025; 17(12): 2090.

[22]Huang X, Wang Q, Li X, et al. Navigation risk assessment of Arctic shipping routes based on Bayesian networks. Journal of Marine Science and Engineering. 2025; 13(12): 2306.

[23]Zhang J, Wu L, Zhang T, et al. Systemic and dynamic risk analysis of drilling construction based on Bayesian network and system dynamics model. Scientific Reports. 2025; 15(1): 27989.

[24]Bakhtiari S, Najafi MR, Goda K, et al. A dynamic Bayesian network approach to characterize multi-hazard risks and resilience in interconnected critical infrastructures. Reliability Engineering & System Safety. 2025; 257(Part A): 110815.

[25]You Q, Guo J, Zeng S, Che H. A dynamic Bayesian network based reliability assessment method for short-term multi-round situation awareness considering round dependencies. Reliability Engineering & System Safety. 2024; 243: 109838.

[26]Li X, Chen R, Ren Y, Jim CY. Intrinsic drivers of urban flood disasters from the resilience perspective in China. Reliability Engineering & System Safety. 2025; 265(Part B): 111596.

[27]Yuan Z, Cai L, Xie Z, et al. Dynamic evolution and scenario-based prediction of urban flood resilience: A system dynamics modeling approach in Kunming, China. Journal of Environmental Management. 2025; 395: 127740.

[28]Reason, JT. Human Error. Cambridge University Press; 1990.