Summary
Monte Carlo simulation is rapidly superceding deterministic methods as the
preferred technique for many well-forecasting applications. This paper
contrasts the two techniques and highlights the potential benefits of the
former. The main part of the paper is a discussion of the application of Monte
Carlo simulation to time and cost estimation of single wells and, in
particular, of some common errors that the authors have observed practitioners
to make (or have themselves made) within this context. This discussion,
together with sections covering multiwell programs, scheduling, and production
issues, demonstrates that although Monte Carlo simulation has the potential to
enhance the reliability of well forecasts, this potential can be realized only
if the technique is applied with great care.
Introduction
Predictions of the duration and cost of well construction, project
schedules, production rates, and cash flows are key inputs into the appraisal,
planning, and monitoring of wells projects. These predictions may collectively
be termed well forecasts. Some of these forecasts, while of interest in
themselves, are also used as the building blocks of other forecasts. Thus,
there emerges a hierarchical structure (Fig. 1).
The sustained delivery of new production at reasonable cost and the
achievement of results signaled to and/or expected by investors are central to
an E&P company’s profitability and reputation. Errors in well time and cost
forecasts will propagate through to production and cash-flow forecasts,
potentially putting the company’s entire business plan at risk. If the
forecasts are basically sound, random variation caused by the inherent
unpredictability of well operations ought not to have a significant effect on
the final, rolled-up results. In contrast, the effects of systematic errors in
forecasting technique are likely to be cumulative, resulting in possible
substantial variance between forecast and actual results. It is the
identification and elimination of these errors in technique with which this
paper is concerned.
© 2006. Society of Petroleum Engineers
View full textPDF
(
1,348 KB
)
History
- Original manuscript received:
7 June 2004
- Revised manuscript received:
16 January 2006
- Manuscript approved:
11 March 2006
- Version of record:
20 September 2006
View Cart
