Hedging against Uncertain Future Development Plans in Closed-loop Field Development Optimization

Abstract

Abstract Optimization has received considerable attention in oilfield development studies. A major difficulty is related to handling the uncertainty that can be introduced from different sources. Traditionally, geologic uncertainty has been considered as an important source of risk and stochastic approaches have been developed to incorporate the related uncertainties in optimization problems. An important aspect to consider in field development optimization is the possibility of future developments (e.g. infill drillings). Disregarding future development activities in optimization of current decisions can lead to field performance predictions and optimization results that may be far from optimal. Recent studies have focused on optimizing future development activities by including them as decision variables in optimization frameworks. A main issue with this approach is that the optimization results are rarely implemented in future developments exactly as obtained, for various reasons including uncertainty in models, drilling and geological considerations, and unpredictable circumstances that directly affect reservoir management and future development decisions. Therefore, it is more practical to consider future development plans as uncertain events that should be hedged against during optimization. In this paper, we show the importance of considering future development plans with their associate uncertainty in closed-loop oilfield development optimization. We present a stochastic closed-loop field development optimization formulation to account for geologic uncertainty and the uncertainty in future infill drilling, where model-based optimization and data integration loops are repeated through reservoir’s life cycle. In the new stochastic formulation, future development events are modelled as uncertain parameters that lead to multiple possible development scenarios. Stochastic optimization formulation finds optimal solutions for current decision variables (e.g. well locations and operational settings) while accounting for the uncertainty in geologic description and future development plans for the remainder of the reservoir life. Thereafter, the reservoir is operated based on optimal solutions for a period of time while reservoir response data is collected and used to calibrate the reservoir models and reduce the geologic uncertainty. The optimization process for current decisions is repeated after each model calibration step to include the new information in decision making and reservoir operation. Once a drilling decision is made, the optimization process is performed with the most recently updated reservoir models, while considering further future development scenarios as uncertain parameters. The uncertain parameters representing the future development plans include the number of future infill wells. A stochastic process is used to describe the uncertainty in the parameters over reservoir development stages, resulting in a decision tree representation with multiple development scenarios. A robust closed-loop optimization workflow is developed to optimize current decision variables by considering plausible geologic model realizations and possible future development scenarios to hedge against the uncertainty they represent. Case studies are presented to illustrate the importance of considering future development plans in closed-loop oilfield development optimization. The performance of the developed stochastic framework is evaluated and discussed relative to existing approaches that do not account for future field development plans in the optimization procedure. The results show significant improvements in the production performance when future field development planning is considered. We observe that incorporating the uncertainty in future development offers flexibility and robustness to accommodate alternative development options, and avoids solutions that significantly underperform when reservoirs undergo unanticipated development (drilling) activities in the future. This study presents the first attempt to consider the uncertainty in future development plans in closed-loop field development optimization. The developed method offers a novel perspective and framework for formulating and solving production optimization problems under geologic and future development uncertainty.