TY - GEN
T1 - A Multi-Layer Heterogeneous Network Risk Identification Framework for Aero-Engine Supply Chains Based on Mutual Information and Cascading Failure Analysis
AU - Wang, Zeyu
AU - Meng, Xueyu
AU - Zhao, Haoxiang
AU - Cai, Zhiqiang
N1 - Publisher Copyright:
© 2026 IEEE.
PY - 2026
Y1 - 2026
N2 - Aero-engine supply chains are typically characterized by complex supplier hierarchies, heterogeneous participating entities, coupled evaluation indicators, and hidden risk transmission. In these networks, risks are usually not directly triggered by a single indicator. Instead, they are associated with, and influenced by, different layers, entities, and processes. Traditional identification methods based on expert experience, single-indicator thresholds, and linear correlations have difficulty revealing nonlinear relationships and cross-layer risk propagation paths in supply chains. Therefore, this paper proposes a multilayer heterogeneous risk identification method for aero-engine supply chains based on mutual information and cascading failure analysis. First, a set of supply chain indicators is used as network Nodes. Normalized mutual information is used to calculate association Strength between indicators, and an undirected weighted network is constructed after TMFG-based sparsification. Second, an indicator-risk mapping scheme is established through the SCOR model, linking existing indicator data with risks in the business processes of Plan, Source, Make, Deliver, and Return. Third, key sentinel indicators are identified based on network centrality measures. The selected indicators may play important roles in risk propagation. Finally, a nonlinear load-capacity model is used to conduct cascading failure simulations. Changes in network robustness under random attacks and targeted attacks are compared, and risk disposal levels are formed according to the influence of Node failures. The application analysis shows that this procedure can identify key Nodes with structural importance and propagation influence from the complex indicator system of aero-engine supply chains, providing methodological support for risk detection, early warning, and structural robustness improvement in complex equipment supply chains.
AB - Aero-engine supply chains are typically characterized by complex supplier hierarchies, heterogeneous participating entities, coupled evaluation indicators, and hidden risk transmission. In these networks, risks are usually not directly triggered by a single indicator. Instead, they are associated with, and influenced by, different layers, entities, and processes. Traditional identification methods based on expert experience, single-indicator thresholds, and linear correlations have difficulty revealing nonlinear relationships and cross-layer risk propagation paths in supply chains. Therefore, this paper proposes a multilayer heterogeneous risk identification method for aero-engine supply chains based on mutual information and cascading failure analysis. First, a set of supply chain indicators is used as network Nodes. Normalized mutual information is used to calculate association Strength between indicators, and an undirected weighted network is constructed after TMFG-based sparsification. Second, an indicator-risk mapping scheme is established through the SCOR model, linking existing indicator data with risks in the business processes of Plan, Source, Make, Deliver, and Return. Third, key sentinel indicators are identified based on network centrality measures. The selected indicators may play important roles in risk propagation. Finally, a nonlinear load-capacity model is used to conduct cascading failure simulations. Changes in network robustness under random attacks and targeted attacks are compared, and risk disposal levels are formed according to the influence of Node failures. The application analysis shows that this procedure can identify key Nodes with structural importance and propagation influence from the complex indicator system of aero-engine supply chains, providing methodological support for risk detection, early warning, and structural robustness improvement in complex equipment supply chains.
KW - aero-engine supply chain
KW - complex network
KW - multi-layer heterogeneous network
KW - risk
UR - https://www.scopus.com/pages/publications/105052994936
U2 - 10.1109/SRSE70977.2026.11695006
DO - 10.1109/SRSE70977.2026.11695006
M3 - 会议稿件
AN - SCOPUS:105052994936
T3 - 2026 the 8th International Conference on System Reliability and Safety Engineering, SRSE 2026
SP - 442
EP - 447
BT - 2026 the 8th International Conference on System Reliability and Safety Engineering, SRSE 2026
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 8th International Conference on System Reliability and Safety Engineering, SRSE 2026
Y2 - 7 August 2026 through 10 August 2026
ER -