Fault and event tree analyses for process systems risk analysis: uncertainty handling formulations

Area of Science:

• Process safety engineering
• Risk assessment methodologies
• Uncertainty quantification

Background:

• Quantitative risk analysis (QRA) is crucial for evaluating process system safety.
• Traditional event tree analysis (ETA) and fault tree analysis (FTA) rely on crisp probabilities and independence assumptions.
• These assumptions often fail to capture real-world uncertainties and interdependencies.

Purpose of the Study:

• To address limitations in traditional QRA by incorporating uncertainty and dependency.
• To develop a robust QRA framework for process systems using advanced uncertainty theories.
• To improve the accuracy and realism of risk assessments in industrial settings.

Main Methods:

• Application of fuzzy set theory and evidence theory to model uncertainties in input event likelihoods.
• Development of a dependency coefficient method to quantify interdependencies between events in ETA and FTA.
• Validation of the proposed approach through two practical case studies.

Main Results:

• Fuzzy set and evidence theories effectively describe uncertainties in event likelihoods, overcoming limitations of crisp probabilities.
• The dependency coefficient accurately captures interdependencies, enhancing the realism of ETA and FTA.
• Case studies demonstrate the practical applicability and improved accuracy of the enhanced QRA framework.

Conclusions:

• The proposed method provides a more realistic and comprehensive approach to QRA for process systems.
• Handling uncertainties and dependencies is vital for accurate risk evaluation and safety management.
• This research offers a valuable tool for improving safety and decision-making in complex industrial environments.