Modeling NaTech-related domino effects in process clusters: A network-based approach

Meng Lan, Paolo Gardoni, Rongshui Qin, Xiao Zhang, Jiping Zhu, Siuming Lo

Research output: Contribution to journalArticlepeer-review


Extreme disasters and their resulting cascading accidents are always being a considerable threat to process cluster safety. However, many current assessments of the domino effects within process clusters do not consider the uncertainty of the primary accidents triggered by natural hazards. Additionally, high-efficiency methods to model the complex dependencies of large-scale process clusters are still lacking. Hence, this paper expands a hazard scenario module at the front end of the assessment framework of a domino effect to consider the response of installations to hazard loads. Simultaneously, a network-based approach is developed to model the NaTech-related domino effect. The constructed network imposes two constraints, escalation and probability thresholds, to reduce the computational complexity of traversing potential propagation pathways. As a result, the proposed method can be applied to large-scale process clusters. The safety analysis of an oil storage base subject to hurricanes and concurrent flooding shows that the proposed method can clarify the role of each unit in a local and cross-community domino effect at the node and community levels, respectively. The results inform the implementation of emergency responses to accidents. Furthermore, sources of uncertainty indicate that the network structure of the NaTech-related domino effect is sensitive to hazard intensity.

Original languageEnglish (US)
Article number108329
JournalReliability Engineering and System Safety
StatePublished - May 2022


  • Domino effect network
  • Escalation vector
  • Large-scale process clusters
  • NaTech
  • Process safety

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Industrial and Manufacturing Engineering


Dive into the research topics of 'Modeling NaTech-related domino effects in process clusters: A network-based approach'. Together they form a unique fingerprint.

Cite this