As SPEE Monograph #3 notes, few guidelines exist to evaluate unconventional resource plays. The significant variability in well performance that is inherent to resource plays may preclude the use of an analogous "type" well when evaluating undrilled locations, as is done for conventional reserve evaluation. Yet the necessity of having reliable production forecasts as the foundation for sound investment remains.

The paper uses Monte Carlo simulation to generate an aggregated production forecast distribution for multiple undrilled locations from the forecast distributions for the individual locations. From the central limit theorem, the sum of individual production forecasts tends towards the average value for the undrilled well population times the number of locations as the number of drilling locations becomes very large (infinite).

However, in the real world the number of drilling locations is finite and limited so a forecast based on the "average" for the entire population likely overestimates the results for the group. The reason is that well production rates are characterized by skewed (e.g. lognormal) distributions so more than 50% of the wells drilled will produce at less than the average well rate while a few "stars" will perform at rates that are well above the average.

Accordingly, the first wells in a random drilling sequence will likely produce at below average rates. The group rate will continue to lag the population average with each successive well drilled until "luck" intervenes and a location that produces at well above the average is drilled. Consistently overestimating producing rates as a field is developed can lead to marginal economics or even "gambler’s ruin" if the "star" well is not drilled before the entire drilling budget is spent.

To overcome the practical pitfalls of the central limit theorem, Monte Carlo simulation is used to quantify the uncertainty in the production forecasts of a group of wells. A 10-location drilling example illustrates the variance in the forecasts over a 20-year project life. The resulting P90, P50 and P10 production forecasts provide a means to assess the impact of our uncertain knowledge on production rates, cash flows and project economics.

You can access this article if you purchase or spend a download.