177

Pipelines play a critical role in transporting fluids, and the assessment and mitigation of associated risks are paramount. Over the past decades, pipeline risk models have evolved from empirical, index-based approaches to more quantitative, data-driven models, providing insightful results for operators to make decisions regarding inspection, mitigation, repair, or replacement efforts. However, data uncertainty and limited data availability have consistently posed challenges in quantitative or probabilistic risk assessment models. Therefore, efficiently improving data quality and completeness becomes necessary, which requires a judicious decision on allocating limited resources for data collection efforts. To meet this goal, a sensitivity assessment on pipeline risk models becomes critical.

This research focuses on identifying and understanding the sensitivity of crucial input factors spanning from pipeline specifications to environmental conditions. A data-driven probabilistic risk assessment model was selected, and the baseline parameters were set up based on a selected pipeline. By systematically varying the model parameters in a reasonable range, and observing their impact on the likelihood of failure, the study aims to prioritize factors significantly influencing the likelihood of failure. After determining the range of the selected factors, a Monte Carlo simulation up to a million iterations was conducted with random sampling inside the defined distribution, and the correlation coefficients between these factors and the risk results were analyzed. As a result, a ranking of the criticality of these factors is summarized. Additionally, a threat-based sensitivity analysis was conducted, specifically on pipeline threats such as corrosion, mechanical damage, incorrect operation, etc., to evaluate which factor affects the likelihood of a specific threat more.

 The results contribute to informed decision-making and resource allocation for risk mitigation strategies. Additionally, the research emphasizes the importance of iterative sensitivity analyses, considering uncertainties, and continuous refinement of the risk assessment model to enhance its accuracy and relevance in dynamic operational environments. The findings of this study offer valuable insights for stakeholders involved in pipeline management, safety, and regulatory compliance.

Read/Download PaperGradeAvailable after the event