Volume 22, Issue 1 (2025)
ioh 2025, 22(1): 0-0 |
Back to browse issues page
Ethics code: IR.IUMS.REC.1403.028
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Keyvani Brojeni M, Kheirandish Sarabi S, Vosoughi S, Souri B, Akbarzadeh Miandoab F, Moradi Hanifi S. QUANTITATIVE RISK ASSESSMENT OF A CARBON DIOXIDE STORAGE TANK BY USING CONSEQUENCE MODELING AND FUZZY BAYESIAN NETWORKS. ioh 2025; 22 (1) : 12
URL: http://ioh.iums.ac.ir/article-1-3735-en.html
URL: http://ioh.iums.ac.ir/article-1-3735-en.html
Mohammadhosein Keyvani Brojeni 
, Somayeh Kheirandish Sarabi , Shahram Vosoughi , Behzad Souri , Fatemeh Akbarzadeh Miandoab , Saber Moradi Hanifi *
, Somayeh Kheirandish Sarabi , Shahram Vosoughi , Behzad Souri , Fatemeh Akbarzadeh Miandoab , Saber Moradi Hanifi *
Occupational Health Research Center Department of Occupational Health Engineering, School of Public Health, Iran University of Medical Sciences , moradi.sab@iums.ac.ir
Abstract: (282 Views)
ABSTRACT
BACKGROUND AND AIMS: Given the asphyxiant and odorless nature of carbon dioxide gas, which, while non-toxic, can pose a serious threat to human life at high concentrations, this study aims to conduct a more precise risk assessment. Since the conventional Bowtie method, by itself, is incapable of dynamic analysis and does not adequately account for uncertainties, this research integrates the causal framework of the Bowtie method with Bayesian Networks, together with consequence modeling and risk assessment using SAFETI 8.4 software. This integrated approach provides a practical framework for enhanced hazard identification, more accurate evaluation of CO₂ leak consequences, and improved preparedness against potential accidents.
METHODS: In this study, after hazard identification using the HAZOP method, leakage scenarios were defined and analyzed through an integrated Bowtie–Bayesian Network approach implemented in GeNIe 5.0. Event probabilities were estimated using a fuzzy logic-based method, and the consequences of leakages were modeled with PHAST 8.4. Finally, risk assessment was conducted using the probit equation in SAFETI 8.4.
RESULTS: In the BOWTIE approach, a total of 17 basic events and 13 intermediate events were identified. The probability of the final consequence resulting from CO₂ leakage, calculated using the Bowtie method and the integrated Bowtie–Bayesian Network approach, was 7.832726×10⁻⁶ using the BOWTIE approach and 7.305449×10⁻⁶ when combined with the Bayesian Network. The most critical basic events were identified as: (1) failure in equipment maintenance, (2) failure in connections and clamp No. 1, and (3) poor quality of purchased equipment. The highest concentration in the event of catastrophic rupture was determined to be 106 ppm at a distance of 5 meters from the tank. The risk level was less than 10-4, which is within the acceptable level.
BACKGROUND AND AIMS: Given the asphyxiant and odorless nature of carbon dioxide gas, which, while non-toxic, can pose a serious threat to human life at high concentrations, this study aims to conduct a more precise risk assessment. Since the conventional Bowtie method, by itself, is incapable of dynamic analysis and does not adequately account for uncertainties, this research integrates the causal framework of the Bowtie method with Bayesian Networks, together with consequence modeling and risk assessment using SAFETI 8.4 software. This integrated approach provides a practical framework for enhanced hazard identification, more accurate evaluation of CO₂ leak consequences, and improved preparedness against potential accidents.
METHODS: In this study, after hazard identification using the HAZOP method, leakage scenarios were defined and analyzed through an integrated Bowtie–Bayesian Network approach implemented in GeNIe 5.0. Event probabilities were estimated using a fuzzy logic-based method, and the consequences of leakages were modeled with PHAST 8.4. Finally, risk assessment was conducted using the probit equation in SAFETI 8.4.
RESULTS: In the BOWTIE approach, a total of 17 basic events and 13 intermediate events were identified. The probability of the final consequence resulting from CO₂ leakage, calculated using the Bowtie method and the integrated Bowtie–Bayesian Network approach, was 7.832726×10⁻⁶ using the BOWTIE approach and 7.305449×10⁻⁶ when combined with the Bayesian Network. The most critical basic events were identified as: (1) failure in equipment maintenance, (2) failure in connections and clamp No. 1, and (3) poor quality of purchased equipment. The highest concentration in the event of catastrophic rupture was determined to be 106 ppm at a distance of 5 meters from the tank. The risk level was less than 10-4, which is within the acceptable level.
CONCLUSION: The social risk curve showed that the risk associated with the carbon dioxide tank is at an acceptable level, indicating a low level of risk. Complex models, such as Fuzzy Bayesian Networks (FBN), can significantly contribute to optimal decision-making in safety and crisis management, particularly when modeling high-risk accidents and assessing associated risks.
Article number: 12
Type of Study: Applicable |
Subject:
Safety
Received: 2025/03/6 | Accepted: 2025/07/8 | Published: 2025/03/30
Received: 2025/03/6 | Accepted: 2025/07/8 | Published: 2025/03/30
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
