LAUSR

Abstract Numerous oil and gas wells in the North Sea are now reaching their end of productive life. The main challenges at the end of the well life phase are: (i) how to recover the remaining resources, and (ii) how to plug and abandon (P&A) the wells and associated structures in a cost-efficient way. Well P&A operations could according to a study contribute with up to 25% of the total costs of exploration wells in the North Sea. The long-term sealing capability will also be a critical factor because of substantial costs of potentially having to re-enter a leaking wellbore. Probabilistic risk assessments (PRA) will be key in identification of optimal P&A solutions. The use of PRA in cost and risk control of well systems are not new, however, the amount of research, guidelines and frameworks proposed for PRA of long-term well P&A concepts remain scarce. A literature search revealed only a few methods proposed that may be of some direct relevance to PRA. Most of the available literature is related to leakage risk modelling associated with CO2 capture and storage, which as such may not be directly adoptable to North Sea exploration wells. In this article we: (i) identify and discuss key risk factors for developing a PRA modelling framework for well systems in the P&A phase on the Norwegian Continental Shelf (NCS), and (ii) based on Weibull statistics and a strongest link argument, propose a novel modelling approach to PRA within this framework.