A hybrid Monte Carlo and graph-based method for analyzing LOC-induced domino effect evolution in the chemical industry

Abstract

Domino effects triggered by loss of containment (LOC) events in chemical facilities represent critical challenges to industrial safety due to their complex, multi-phase, and spatially evolving characteristics. Traditional assessment methods often fall short in capturing both the dynamic propagation of accidents and the systemic vulnerabilities within facility layouts. To address these limitations, this study proposes a hybrid analytical framework that integrates time-dependent Monte Carlo simulations with graph-theoretic metrics to model and analyze the spatiotemporal evolution of LOC-induced domino effects. Unlike conventional static models, the proposed approach explicitly incorporates fire–explosion coupling and dynamic unit-state transitions, enabling high-resolution simulation of cascading failures over time. The framework is applied to the CAPECO tank farm explosion as a representative case study, demonstrating strong consistency with the observed progression of the real-world accident. The results provide actionable insights for risk-informed facility layout, hazard mitigation, and emergency response planning. Overall, the proposed method offers a significant advancement in the modeling, understanding, and management of domino effects in complex chemical storage environments.