Reliability Modeling of Fault Propagation in Software-Hardware Coupled Intelligent Ship Systems Using Petri Nets and Embedded Dynamic Bayesian Networks

Luo, Xiaofang and Guo, Linghui and Ma, Xiandong and Bai, Xu and Li, Jingling (2026) Reliability Modeling of Fault Propagation in Software-Hardware Coupled Intelligent Ship Systems Using Petri Nets and Embedded Dynamic Bayesian Networks. Reliability Engineering and System Safety: 112597. ISSN 0951-8320 (In Press)

Text (MS 20260309_accepted)
MS_20260309_accepted.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (1MB)

Abstract

Maritime autonomous surface ships are increasingly recognized as complex cyber-physical systems, where autonomous navigation depends on tightly coupled software and hardware components. Conventional reliability assessment approaches often assume independence between software and hardware, which limits their accuracy in capturing fault propagation mechanism. This paper proposes a novel reliability modeling framework that explicitly incorporates software-hardware coupling. First, Fault propagation paths are represented using Petri Nets (PN) to capture dynamic interactions among components. The PN structure is then transformed into an embedded Dynamic Bayesian Network (DBN), enabling dynamic probabilistic reasoning over fault dependencies. By integrating PNs with DBNs, a coupled reliability model is constructed for intelligent ship autonomous navigation systems. A case study on an autonomous navigation task demonstrates the effectiveness of the proposed method in quantifying system reliability, identifying critical software and hardware components, and highlighting fault propagation effects that are overlooked in a decoupled model. The results confirm that the proposed approach enhances both the accuracy and interpretability of reliability assessment for intelligent ship systems.