Quantifying the reliability of surface bops with subsea shut-off assembly and sub-surface bops, taking into account human reliability

Selecting a reliable blowout preventer system is a challenging decision in deepwater drilling. For many years, subsea BOP has been used as the only available technology for drilling deepwater wells. Recently, offshore drilling companies have applied a new concept for blowout control for wells up to 10,000 ft. deep. They proposed using surface BOP with two extra shear rams installed at the seabed. This additional safety barrier is called subsea shut-off assembly (SSA). In this research we calculated the blowout probability using each BOP configuration. To consider human and organization factors as part of the BOP system, we used the human reliability analysis method SPAR-H. We used Fault Tree Analysis (FTA) to graphically present the equipment and human causes of failure. Monte Carlo simulation was used for stochastic calculation of blowout probability. In this approach, we model the system failure with numerous realizations based on the individual human or equipment failure probability distributions. This resulted in a more accurate representation of the system compared to deterministic analysis. Comparing reliability of two BOP configurations shows that surface BOP with SSA is more reliable than subsea BOP. However, surface BOP with SSA has operational limitations in harsh metocean conditions. Therefore, selecting the BOP configuration depends on operational factors as well as reliability. We examined the sensitivity of BOP system to equipment and human failures. Based on the sensitivity analysis, we recommended cost-effective modifications for each BOP system to increase the reliability. To evaluate the human error contribution, we calculated the blowout probability by removing all the human causes from the fault trees. The blowout probability without human error is one order of magnitude smaller than the original blowout probability. Therefore, ignoring the human failure in BOP reliability studies would results in significant underestimation of the blowout risk.