Hazard identification and Dynamic risk assessment in vulnerable areas of a gas compressor station

Document Type : Research Article


1 Iran University of Science and Technology (IUST)

2 Iran university of science and technology (IUST)



In this study, a systematic approach is introduced for process hazard Identification and dynamic risk assessment focusing on the process equipment failure as well as control and instrument facilities in a typical gas compressor station. The survey was initially performed via HAZOP analysis for the main process facilities of the case study. Then, the Bow-Tie analysis was proposed as a combination of fault tree and event tree. Because of the static nature of this conventional risk assessment methods and the limitations such as uncertainty in modeling casual relationships between components, this study proposes a dynamic approach using Bayesian networks to safety risk assessment of gas compressor station. After updating the probability of the main event occurrence using sensitivity analysis, critical points in the unit were identified. The HAZOP results showed that the compressor node had a high potential to loss of containment or compressor failure. The evaluation of human error is also performed using specific data. The results of quantitative risk assessment show that the probability of compressor damage due to surge condition is equal to 2.163 × 10-3. It is also found that probability of jet fire and flash fire, are equal to 1.51E-6 and 9.58E-6. For dynamic risk assessment the bow-tie was mapped to the Bayesian network. After setting the new evidence, the probabilities of intermediate events and basic events were updated. Sensitivity analysis using ROV criterion, showed that compressor protection system failure, anti-surge control valve failure, and the events associated to the compressor strainer were the main contributors.


Main Subjects