A review of risk analysis and accident prevention of blowout events in offshore drilling operations

Offshore drilling operations are complex and hazardous as it is exposed to harsh environment conditions. During drilling process, kick can occur due to low well column pressure and high formation pressure. If kick gets undetected, it can lead to Blowout that has catastrophic effects to man, material and environment. Over past decades, Blowout has been identified as a major cause for many offshore accidents including those caused by equipment malfunction or well control barrier failures. It is therefore crucial to investigate the underlying causes of Blowout events for accident prevention. A review of risk analysis of Blowout occurrences utilizing quantitative methods and accident prevention using machine learning techniques is reviewed in this article. In quantitative risk analysis, Bow-Tie models have made foremost contributions to understanding the causes and consequences of Blowout. Application of Bow-Tie and Bayesian Network have made a paradigm shift in dynamic risk analysis by estimating updated probabilities of drilling accidents. This has helped drilling engineers to pinpoint root causes and degree of influence of major factors contributing to accidents. To prevent Blowout and to predict its initiating events, various machine learning algorithms employing supervised, unsupervised and deep learning techniques have developed. These algorithms could effectively diminish frequency of drilling accidents by facilitating early detection and prediction of accident initiating events. Risk analysis methods and accident prevention algorithms have demonstrated a deeper insight into drilling operations; however, it necessitates the need to extend the scope of work towards advanced facets of drilling operations for effective safety, health and risk management.