A Novel Approach to Identify Reservoir Analogues

Abstract

Abstract The identification of reservoir analogues if often an important step in planning the development of a new play, particularly when it poses difficult technical challenges or when appraisal information is limited. One of the strengths of an E&P company with a large and diverse portfolio is its potential to utilize a broad base of knowledge gained from past experience to aid in future development and decision-making. A critical enabler in being able to leverage this knowledge effectively is the ability to access relevant information quickly and efficiently. This paper presents a new approach to identifying analogues systematically through the novel application of an artificial intelligence technique called case-based reasoning. The key lies in characterising each reservoir by a set of attributes which describe the reservoir and can be used to differentiate it from other reservoirs. Case-based reasoning is then applied to search for reservoirs which are similar (i.e. are potentially good analogues), but that do not necessarily match up exactly on any individual attribute. Further, this method quantifies the degree of similarity and permits users to focus their search on any specified set of reservoir attributes. These characteristics differentiate this method from others currently in use. The implementation of this method to create a versatile knowledge sharing tool, called the "Smart Reservoir Prospector", will be described. Recently developed artificial intelligence technology incorporating case-based reasoning algorithms within a flexible java-based architecture has been applied. The result is a web-based system that permits users in any Shell operating unit to search through thousands of reservoirs in milli-seconds to locate reservoir analogues, and then to access additional detailed information about the reservoirs of interest. The application of this tool within Shell to address important business issues, including scope for recovery and reserves justification, recovery factor benchmarking, and reservoir knowledge sharing, will be discussed. Introduction This article describes the origin and development of a unique knowledge sharing tool for locating reservoir analogues and sharing associated best practices. The tool, called the Smart Reservoir Prospector (SRP), is accessible globally within Shell. Also discussed are the underlying business drivers and the benefits of using reservoir analogues to aid in exploration and production activities. Finally, a vision for the further development and embedment of the SRP within Shell EP's knowledge sharing communities is presented. The SRP is a reservoir knowledge sharing system with an intelligent "fuzzy logic" search facility. Its main purpose is to locate reservoir analogues in a systematic, reliable, and efficient way. This contrasts with the traditional hit-or-miss approaches such as searching through reports or relying on the individual experiences of colleagues. These approaches are generally biased towards assessing similarity on a few attributes only, and compare reservoirs on a one to one basis. The SRP assesses similarity on the basis of a broad set of attributes and compares multiple reservoirs simultaneously.