Application of case-based reasoning for well fracturing planning and execution Andrei Popa * , William Wood Chevron Corporation, CA, USA article info Article history: Received 18 November 2010 Received in revised form 21 July 2011 Accepted 22 July 2011 Available online 9 September 2011 Keywords: Case-based reasoning Hydraulic fracturing Artificial intelligence abstract Over the last two decades, there has been significant activity in the soft computing arena with focus on various computer paradigms such as neural networks, genetic algorithms, and fuzzy logic to more efficiently solve complex engineering problems. Further work concentrated on integrating two or more of these paradigms and led to what is known as “hybrid systems”. The power of the hybrid system relies on the fact that technologies are intended to complement each other and exploit their individual strengths to enhance solution generation. Case-based reasoning (CBR) is another soft computing technology developed to deal with uncertainty, approximate reasoning and exploit knowledge domain. Case-based reasoning, also known as computer reasoning by analogy, is a simple and practical technique that solves new problems by comparing them to ones that have already been solved in the past, thus saving time and money. This paper provides a general framework of case-based reasoning along with a review of the four-step cycle that characterizes the technology (retrieve, reuse, revise and retrain), followed by a specific application to well fracture treatment design, planning and execution. The proposed methodology extracts the relevant historical information recorded during field job execution, utilizes a rule-based system to make adaptations, and then suggests the most appropriate solution for new well fracturing candidates. The technique was tested as a front-end tool using sample data from a tight gas field with significant hydraulic fracturing activity. This simple case demonstrates how case-based reasoning can be applied to improve hydraulic fracturing design, planning and execution of wells, thus significantly increasing the job execution success while avoiding known pitfalls. In addition, the work demonstrates the value of captured “on-site” experience and shows the advantages of using intelligent systems in decision-making. Ó 2011 Elsevier B.V. All rights reserved. 1. Introduction One of the most widely used well completion techniques to increase well productivity is hydraulic fracturing. Due to its high success and great returns, hydraulic fracturing is carried out in almost all new wells drilled in tight gas sands. Moreover, wells are often re-fraced once production declines below economic limits or after a long period of time from initial completion. In large tight gas fields, hydraulic fracturing represents the main field activity and significant data and knowledge has been captured over time. Successfully fracturing and/or re-fracturing a well has been a chal- lenge for engineers since the early-times. Significant work and research was conducted in the last decade using artificial intelli- gence techniques to mine data and build predictive systems that would maximize the output of the wells. The study presented in this paper explores the applicability of case-based reasoning technology as the knowledge domain to not only help with optimum treatment design but considerably improve planning and execution of hydraulic fracturing treatments in a gas field. The system makes use of an available database con- taining all the historical fracturing jobs executed in the last decade in the field. The well fracturing datasets represent a tremendous amount of knowledge that can and should be exploited for better decision-making. Case-based reasoning is an excellent technique to maximize the value of historical data enabling better decisions regarding well planning and execution. The approach uses wells with similar characteristics to recommend the appropriate treat- ment design, assist in the planning stage and advise during execution to avoid job failure. 2. Case-based reasoning Case-based reasoning (CBR) is a basic problem solving technique that uses and adapts the solutions of analogous past problems to solve new problems. Its roots are steeped in human cognitive research from the early 1980s, and the technique has gained trac- tion in the last decade. CBR can be described as a common human * Corresponding author. 9525 Camino Media, Bakersfield, CA 93311, USA. E-mail address: andreipopa@chevron.com (A. Popa). Contents lists available at SciVerse ScienceDirect Journal of Natural Gas Science and Engineering journal homepage: www.elsevier.com/locate/jngse 1875-5100/$ e see front matter Ó 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jngse.2011.07.013 Journal of Natural Gas Science and Engineering 3 (2011) 687e696