ABSTRACT
A risk-based approach has been demonstrated to be capable of evaluating the quality of permanently plugged and abandoned petroleum wells. The quality measure in this context is the leakage risk, expressed in terms of probability of barrier failure and the leakage rate (consequence), where associated uncertainties are dealt with by means of probability distributions. In complex engineered systems, such as a barrier system in a permanently plugged and abandoned oil or gas well, it is reasonable to question whether the probability distributions provide adequate representations of the uncertainties. To improve the risk-based approach for plug and abandonment, and to contribute to more informed decisions by better reflecting the uncertainties upon evaluation of the leakage risk, in this paper, we propose an approach to assess the assumptions made and reflect the strength of knowledge, to complement the probability distributions. An example is included to illustrate the approach.
