A Standard Taxonomy for Risks that Derive from Subsurface Uncertainties

Smalley, Craig; Chebotar, Konstantin
BP, Sunbury on Thames, United Kingdom.

The oil and gas industry is based on finding and producing an inherently uncertain resource. Because oil and gas accumulations can only be sampled partially and remotely, there will always be uncertainties in factors such as the accumulation shape, size, and properties. These uncertainties can lead to risks, where a risk is defined as an "event", i.e. a situation that could occur and, if so, would have a material impact on the business. Such risk events are described in words, using the format cause - event - consequence. For example "the presence of a sealing fault (cause) divides the reservoir into two compartments (event) resulting in lower recovery from the planned well (consequence)".

The management of such risks is greatly facilitated if learnings can be easily transferred between different fields within a company, or between partners. For this, a standard way of describing the risks is essential. Here, we propose a risk taxonomy based on a three-level hierarchy employing the cause-event-consequence structure. Each risk is first assigned a Level 1 classification, based on the primary area of consequence (e.g. HSE, in-place volumes, recovery factor, production rate, life-of-field integrity, operating efficiency, product quality, schedule, cost, relationships). The compartmentalization situation described above would fall in the recovery factor Level 1 category. Next, the risk is assigned to a Level 2 category based on the generic risk event type, e.g. compartmentalization in the above example. Finally the risk is assigned to a Level 3 category based on the subsurface root cause, for example the presence of sealing faults. One advantage of this hierarchy is it always allows risks to be traced back to their root causes; and it allows subsurface uncertainties to be assessed in terms of their possible business consequences.

Using this scheme we identified 11 Level 1 categories, ~100 at Level 2, and a few hundred at Level 3. A preliminary analysis shows that, when risk are classified in this way, it is possible to reveal risk trends in different field types and different stages of field life that were hitherto difficult to quantify.

AAPG Search and Discovery Article #90155©2012 AAPG International Conference & Exhibition, Singapore, 16-19 September 2012