Please use this identifier to cite or link to this item:
Type: Artigo
Title: Stochastic modeling and analysis of vapor cloud explosions domino effects in chemical plants
Author: Sierra, Diego
Montecchi, Leonardo
Mura, Ivan
Abstract: Because of the substances they process and the conditions of operation, chemical plants are systems prone to the occurrence of undesirable and potentially dangerous events. Major accidents may occur when a triggering event produces a cascading accident that propagates to other units, a scenario known as domino effect. Assessing the probability of experiencing a domino effect and estimating the magnitude of its consequences is a complex task, as it depends on the nature of the substances being processed, the operating conditions, the failure proneness of equipment units, the execution of preventive maintenance activities, and of course the plant layout. In this work, we propose a stochastic modeling methodology to perform a probabilistic analysis of the likelihood of domino effects caused by propagating vapor cloud explosions. Our methodology combines mathematical models of the physical characteristics of the explosion, with stochastic state-based models representing the actual propagation among equipment units and the effect of maintenance activities. Altogether, the models allow predicting the likelihood of major events occurrence and the associated costs. A case study is analyzed, where various layouts of atmospheric gasoline tanks are assessed in terms of the predicted consequences of domino effects occurrence. The results of the analyses show that our approach can provide precious insights to support decision-making for safety and cost management
Subject: Modelos estocásticos
Avaliação de riscos
Custos e análise de custo
Country: Inglaterra
Editor: Springer
Rights: Aberto
Identifier DOI: 10.1186/s13173-019-0092-8
Date Issue: 2019
Appears in Collections:IC - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
2-s2.0-85074253129.pdf2.41 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.