Loading...
Reliability Assessment of IoT-enabled Systems using Fault Trees and Bayesian Networks
Abdulhamid, Alhassan Ghafir, Ibrahim Lei, Ci
Abdulhamid, Alhassan
Ghafir, Ibrahim
Lei, Ci
Publication Date
2024-01
End of Embargo
Supervisor
Rights
Peer-Reviewed
Yes
Open Access status
closedAccess
Accepted for publication
2023-11-24
Institution
Department
Awarded
Embargo end date
Additional title
Abstract
The Internet of Things (IoT) has brought significant advancements in various domains, providing innovative and efficient solutions. However, ensuring the safe design and operation of IoT devices is
crucial, as the consequences of component failure can range from system
downtime to dangerous operating states. Several methods have been proposed to evaluate the failure behaviours of IoT-based systems, including
Fault Tree Analysis (FTA), a methodology adopted from other safetycritical domains. This study integrated FTA and Bayesian Network (BN)
models to assess IoT system reliability based on components’ reliability
data and other statistical information. The integrated model achieved
efficient predictive failure analysis, considering combinations of 12 basic
events to quantify the overall system’s reliability. The model also enables
criticality analysis, ranking basic events based on their contributions to
system failure and providing a guide for design modification in order to
enhance IoT safety. By comparing failure data in FTA and criticality
indices obtained using the BN model, the proposed integration offers a
probabilistic estimation of IoT system failure and a viable safety guide
for designing IoT systems.
Version
No full-text in the repository
Citation
Abdulhamid A, Kabir S, Ghafir I et al (2024) Reliability Assessment of IoT-enabled Systems using Fault Trees and Bayesian Networks. 5th International Conference on Advances in Distributed Computing and Machine Learning (ICADCML). 5-6 Jan, VIT-AP University, India in collaboration with College of Engineering, San Diego State University, USA.
Link to publisher’s version
Link to published version
Link to Version of Record
Type
Conference paper