ABSTRACT
The design of an optimal sand control method and production management is a complex problem due to the simultaneous influence of various factors. Typical effective variables for choosing an optimum sand control method include geological, technical, economical, and expert's experience on similar projects. Some technical factors, which affect the optimum method, are the type of exclusion, gravel size of gravel pack and pre-packed screen, slot width and liner slot length, and productivity index reduction. The situation could be more complicated due to the uncertainty associated with various contributing factors. Therefore, it is crucial to develop a novel approach in order to select the best sand control method with a maximum level of confidence. In this study, to select an optimal sand control method, Multi Criteria Decision Matrix (MCDM) techniques including Analytic Hierarchy Process (AHP), Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and, ELimination and Choice Expressing REality (ELECTRE) are used. To simulate fluid flow, an integrated model of reservoir, well, and surface facility is used based on actual oil field data collected from the south of Iran. Then, Design of Experiment (DOE) and Response Surface Methodology (RSM) are applied to optimize the controllable variables of the best selected sand control method by MCDM. Finally, Monte Carlo Simulation (MCS) is applied to perform sensitivity and uncertainty analysis in order to determine the crucial factors that control net present value (NPV). The results show that the best sand control method based on AHP, TOPSIS, and ELECTRE is the slotted liner. After that, three different methods of pre-packed, gravel pack, and wire wrapped are respectively the most efficient sand control methods based on an average score of all the MCDM techniques. The results also indicate that although the pre-packed screen has the highest NPV, it is not the best sand control method due to the influence of other efficient criteria. The result of sensitivity analysis using MCS in terms of contribution to total variance shows that slot width, slot density, and slot height controls 60.5%, 38.8%, and 0.7% of the NPV variation within the range of factors, respectively.
