 The Quantification of Uncertainties in Production Prediction Using Integrated Statistical and Neural Network Approaches: An Iranian Gas Field Case Study

Document Type : Research Article



Uncertainty in production prediction has been subject to numerous investigations. Geological and reservoir engineering data comprise a huge number of data entries to the simulation models. Thus, uncertainty of these data can largely affect the reliability of the simulation model. Due to these reasons, it is worthy to present the desired quantity with a probability distribution instead of a single sharp value.
For the case-study, numbers of parameters which are believed to contribute largely the uncertainty of Field Gas Production Total are recognized. A sensitivity analysis was done to find the most significant initial parameters. Screening experiments are designed in order to recognize the main factors and the significant interactions of factors that we need to certainly include in the response function. Later, experiments of response surface are designed objective to model the response surface function of Field Gas Production Total. This has been done based on applying two methods, Response Surface Methodology and Artificial Neural Networks. The probability distribution of Total Field Gas Production was then plotted using Monte Carlo simulation.


[1] Abdollahzadeh, A.: “Koma Field’s Full-Field Model Update and History Matching”; IFP-school, France; Final MS Thesis in Reservoir Geoscience and Engineering; 2002.
[2] Chewaroungroaj, J.; Varela, Omar J., SPE; and Lake, Larry W.SPE, The University of Texas at Austin: “An Evaluation of Procedures to Estimate Uncertainty in Hydrocarbon Recovery Predictions”, paper SPE 59449 presented at the 2000 SPE Asia Pacific Conference on Integrated Modelling for Asset Management held in Yokohama, Japan, 25-26 April 2000.
[3] Cox, D.R.; Reid, N.: "The Theory of the Design of Experiments", Chapman & Hall/CRC; 1 edition; June 6, 2000.
[4] Damsleth E., SPE; Hage, Asmund, Norsk Hydro A/S; Volden,Rolf, Norwegian Computing Center: “Maximum Information at Minimum Cost: A North Sea Field Development Study with an Experimental Design”, paper SPE 23139; 1992.
[5] Dejean, J. P., IFP; and Blanc, G., SPE, IFP: “Managing Uncertainties on Production Predictions Using Integrated Statistical Methods”, paper SPE 56696 presented at the 1999 SPE Annual Technical Conference and Exhibition held in Houston, Texas, 3-6 October 1999.
[6] JMP6 Software Manual”; SAS Co; 2005.
[7] Lechner, J. P., SPE; OMV AG, and Zangl., G., SPE,Schlumberger Information Solutions: “Treating Uncertainties in Reservoir Performance Prediction with Neural Networks”,paper SPE presented at the SPE Europec/EAGE Annual Conference held in Madrid, Spain, 13-16 June 2005.
[8] Mason, Robert L.; Gunst, Richard F.; Hess, James L.:"Statistical Design and Analysis of Experiments with Applications to Engineering and Science”; Second Edition, AJOHN WILEY & SONS PUBLICATION; 2003.
[9] NIST/SEMATECH “e-Handbook of Statistical Methods”,http://www.itl.nist.gov/div898/handbook/, 2006.
[10] “Smith, O.J.; Hendry, D.J.; and Crowther, A.R., BP Exploration: “The Quantification and Management of Uncertainty in Reserves”, paper SPE 26056 presented at the Wostorn Regional Meeting held in Anchorago, Alaska,U.S.A., 26-28 May 1993.
[11] “Statistica Software Manual”; Statsoft Co; 2003.