Once a reservoir have been satisfactorily history matched by fitting all production data taken up to the time of the study, the reservoir engineer proceeds with the computation of future reservoir performance under alternative depletion strategies. As part of this forecasting exercise a sensitivity analysis is often performed in an effort to quantify the deviation from the expected production levels should certain important reservoir parameters have a different value than that used in the history matched model. It is this uncertainty in the predicted reservoir production levels that is addressed in this manuscript. By using the statistical properties of least squares estimates, the uncertainty in all estimated reservoir parameters can be readily incorporated into the computations. The proposed methodology is illustrated with a typical example where 95% confidence intervals of individual well production rates as well as of the cumulative overall reservoir production are computed.

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