Identify and handle safety challenges during decommissioning of offshore installations

Master's thesis in Risk managementMost of the offshore oil and gas installations are going towards the cessation of their production life, which means that the decommissioning activity will be increasing in years to come. Decommissioning of the offshore installation is a complex and challenging task. A proper risk management process is needed to identify safety challenges and issues associated with decommissioning activities. In this thesis, some significant safety challenges and issues have been identified. The thesis proposes a risk management process that determines the cause and consequences of each hazard by using Bayesian network. Uncertainty assessment procedures have also been included for the risk analysis results to provide useful information to decision makers. In addition, mitigation techniques for identified hazards have been suggested. In the end, a case study has been carried out to implement and show that proposed risk management process provides a better way to foresee decommissioning safety issues and control them effectively. In this thesis, Shell Leman BH field is used as a case study. The comparison is made between Shell risk control framework and suggested risk management process for particular points like risk definition, risk acceptance criteria, and risk assessment matrix. For these particular points, it is found that the general Shell risk management framework provides a vulnerable mitigation plan as it doesn’t include uncertainty associated with the probability values according to new risk perspective proposed by Aven (2013) and by risk definition of PSA (2016). The proposed risk management process in this thesis applied to identify the hazards for decommissioning of Leman BH field. The analysis procedure results given by proposed process is providing better management and mitigation procedure for the safety issues. The proposed risk management process provides a better decision making as it uses Bayesian network together with uncertainty analysis.