Summary
Spurred by improvements in reliability, cost, and accuracy, sensors offer a
means of increasing expected ultimate hydrocarbon recovery in producing assets
as well as in planned and prospective projects. Ultimate hydrocarbon recoveries
larger than those currently achieved are possible, especially when sensors are
used with advanced recovery methods. However, it is often unclear if the
incremental recovery justifies the cost of installing the sensors. This paper
proposes a method for estimating incremental values attributable to real-time
sensors and provides a demonstration of the method for several production
technologies and reservoir settings. The method offers a transparent and
practical means of making value of information (VOI) computations to be
implemented readily by project teams. An additional benefit of this method is
that the process of specifying the inputs to the analysis facilitates a
systematic discussion of strengths and weaknesses, and builds consensus
regarding assumptions. The method is applied to four scenarios developed by a
panel of industry experts to represent generic, but yet realistic reservoirs.
The results for these scenarios indicate the value of sensors depends on the
market price for product and the type of reservoir and production technology.
The greatest absolute economic value for the use of sensors is obtained for a
deepwater reservoir, while the greatest economic value per equivalent barrel of
oil produced is obtained for a mature onshore reservoir. These expected
economic values are intended to be compared to the cost required to implement
the sensors to assess whether or not there is an expected net benefit.
Introduction
Formal methods of valuing information (sometimes called monetizing
information) have existed in the research and professional literature for many
years. Most publications on VOI have appeared in financial, economic,
operations research, or decision analysis journals (Roberts and Weitzman 1981);
little has appeared in engineering publications, especially petroleum
engineering publications. Recently, a review of VOI in the oil and gas industry
was presented by Bratvold et al. (2007).
VOI methods are simple at the highest conceptual level: the values for
courses of action with and without sensors are estimated and compared. The
difference between the expected values with and without sensors is the expected
value of the sensors and therefore represents the maximum willingness to pay
(WTP) for the sensors. If the WTP for the sensors is greater than the cost of
installation (e.g., sensor cost, installation costs, and deferred production)
and operation of the sensors, their installation is expected to provide a net
benefit.
VOI assessments have the following components:
- They account for uncertainty in the outcome of decisions. The existence of
uncertainty is the reason the valuation is based on expected values.
- They capture the ability of the sensors to change a decision. Typical
decisions are an optimization of the current technology, immediate changes in
technology, or the nature and timing of future technology changes.
- They allow for the sensors to change the monetary outcome of a course of
action even when a decision is not changed by the information.
This paper proposes a method for VOI assessment of real-time sensors and
demonstrates the method for four different combinations of hydrocarbon recovery
technologies and reservoir settings: (1) CO2 injection in mature oil
reservoirs, (2) steam-assisted gravity drainage in heavy oil reservoirs, (3)
hydraulic fracturing in tight gas reservoirs, and (4) waterflooding in
deepwater sandstone reservoirs. Drawing on industry experts, significant effort
was made to make the cases as realistic as possible so the results can be used
to inform the development of project- and corporate-level plans regarding the
use of sensors. But, because of project and portfolio idiosyncrasies, the
results are not to be viewed as definitive or totally generalizable.
© 2009. Society of Petroleum Engineers
View full textPDF
(
573 KB
)
History
- Original manuscript received:
1 March 2007
- Meeting paper published:
11 April 2007
- Revised manuscript received:
16 February 2009
- Manuscript approved:
25 February 2009
- Published online:
31 July 2009
- Version of record:
9 September 2009
View Cart
