Agent-based model and simulation of mitigated domino scenarios in chemical tank farms

Federica Ovidi, Laobing Zhang, Gabriele Landucci*, Genserik Reniers

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

6 Citations (Scopus)


The growing trend of interconnecting two or more chemical process or storage facilities represents a critical safety issue, since an accident can easily escalate from an industrial establishment to the nearby plants resulting in a domino effect. However, common safety analyses often ignore cascading events in chemical tank farms, their complex and transient evolution, and mitigation effects of add-on safety measures. The aim of the present work is to develop a structured approach for the assessment of complex domino events accounting for the influence of safety barriers. The approach is based on the adoption of Agent-based Model and Simulation for the assessment of Domino effect in presence of add-on Protections (DAMS-P). For the first time, the assessment of mitigated cascading events in chemical tank farms is carried out accounting for the transient evolution of multiple scenarios and related synergistic effects, and the effect of safety barriers and their possible time-dependent degradation. A verification of DAMS-P is firstly performed through the comparison against analytic probability evaluation based on event tree analysis and tested through the application of industrial cases. The results obtained constitute a useful support for decision-making and for the identification of critical barriers and their performance evaluation.

Original languageEnglish
Article number107476
JournalReliability Engineering and System Safety
Publication statusPublished - 2021


  • Agent-based modelling
  • Chemical tank farm
  • Computational experiments
  • Domino effect
  • Process safety
  • Safety barriers


Dive into the research topics of 'Agent-based model and simulation of mitigated domino scenarios in chemical tank farms'. Together they form a unique fingerprint.

Cite this