Today's decision makers in the petroleum industry have access to remarkable new technologies, huge amounts of information to help them make high-quality decisions, and the ability to share information at unprecedented speeds and quantities. A central hypothesis in the concept of digital oil field is that these tools and resources should lead to better decisions. Yet, the tools bring with them daunting new problems: the available data, though massive, are uncertain and often incomplete, unreliable, or distributed; interoperating/distributed decision makers and decision-making devices need to be coordinated; and many sources of data need to be fused to yield a good decision. In this paper, we shall illustrate the use of Influence Diagrams, also known as Bayesian Decision Networks, to frame, analyze, and support real-time drilling decisions. Influence diagrams provide compact graphical representational and computational frameworks that are based on rigorous probability theory. They consist of a framing part, which encodes the decision basis elements along with their interrelationships, and a probabilistic part, which uses conditional probability distributions to quantify the strengths of the relationships. Following a systematic decision analytic framework, particularly the decision analysis cycle, examples of influence diagrams are developed to provide insight to the decision situations. Also, an object-oriented influence diagram is built and tested for a real world casing setting decision situation. The influence diagram models are updated with the arrival of new data to reflect the current state of knowledge about the decision situation. The study shows that influence diagrams are well-suited and efficient decision analysis tools for real-time support of the large and complex drilling decisions in which uncertainty is predominant. Introduction Drilling costs account for 50% of the upfront expenditures in oil & gas field development (Saputelli et al., 2003). As operators drill for oil and gas in increasingly complex and challenging environments, drilling operations become increasingly costly and risky. In the current economic environment, improving the drilling process is acknowledged to facilitate increased production and reserves while reducing costs. The industry has approached this challenge in two ways:by improving the productive time of the routine drilling process itself; andby eliminating the additional drilling expenses due to non-productive time related to unexpected and undesired drilling events.
One of the objectives of Integrated Operations (IO) 1 is to automate the decision-making process in drilling operations. Such automation is essential to improve both decision quality and speed in the complex and data-rich environment of drilling operations. Over the past decade, the E&P industry has significantly improved its ability to acquire and share large data volumes at unprecedented speeds. The motivation for these investments is to increase value through increased future productions while simultaneously reducing drilling costs. However, utilizing the large data volumes is not always straightforward. Sensor measurements are uncertain because of their reading errors and possible malfunctions, and many sources of data need to be fused to provide decision-relevant information. Furthermore, many drilling operational decisions must be made with very limited time, and this makes evaluating potential alternatives based on the decision criteria by the geosteering team difficult or even impossible. Given these challenges, there is a need to develop efficient processes and tools to automatically support drilling operational decisions. In this paper, we propose a modular intelligent decision advisor (IDA) framework for supporting geosteering decisions. The first module validates and combines the sensor data to determine the probability of the sensor faults and failures as well as undesired geosteering events. These events may include a dramatic increase in drag friction factors, approaching bed boundaries, tools damage, etc. Failure to detect these events could lead to losing the steering ability and sometimes catastrophic problems. The second module proactively analyzes geosteering hazards and recommends the best alternatives. To perform the sensor validation/fusion and hazard/decision analysis, we suggest the use of influence diagrams, also known as Bayesian decision networks (BDNs). Developing an IDA can lead to the reduction of human operators from rig-sites, drawing its value from improving the human safety and geosteering efficiency. Implementing IO with its associated technologies such as large volume data transmission and computational capabilities provides the means for this autonomous decision advisor. Our results show that the BDN is a useful and relevant element in the autonomous geosteering decision-making process. IntroductionHorizontal drilling is well established as a cost effective solution for exploiting oil and gas in challenging environments such as deep water and thin formations. To place the wellbore in an optimal trajectory, the directional driller adjusts the drill-bit position (inclination and azimuth) to reach one or more geological targets. These adjustments are based on the analysis and interpretation of real-time sensor readings and subsurface model outputs, combined with subject-experts' knowledge in the team. The success of geosteering operations, therefore, relies on timely access to relevant decision-supporting information. To optimize drilling operations, operators have gradually inc...
A systematic decision analytic methodology for providing insight into horizontal well placement decisions is presented. These decisions require access to real-time data and to input from subject-experts on the team and from the decision-maker. The available information and outcomes of alternatives are associated with uncertainty. Furthermore, the decision-maker may have multiple objectives, such as future productions, wellbore configuration, and drilling costs. The methodology uses Bayesian decision networks, also known as influence diagrams, to frame, analyze, and support operational decisions.Influence diagrams are compact graphical representations of decisions and are based on decision and probability theories. They rigorously illustrate the interrelationships among decisions, key uncertainties, and value functions. Furthermore, they allow incorporating sensor readings, sub-surface model outputs, and experts' knowledge, which enables probabilistic inference and decision support in real-time contexts. Finally, the influence diagrams models may be used to calculate value of information and of control.In this paper, we focus on supporting well placement decisions involving multiple objectives. We explore the feasibility and suitability of using two potential extensions of the influence diagrams: (1) multiple-attribute utility influence diagram and (2) multiple-objective influence diagram. Using the methodology, the example shows that both extensions have the potential to improve the decision-making process within the multi-disciplinary team. They can provide further insight into decisions by ranking the value of information for key uncertainties from the perspective of different sub-groups of the drilling team.Following a systematic decision analytic methodology, in particular, decision analysis cycle, and applying influence diagrams in drilling operations would create value by reducing losses (materials and drilling/completion processes' downtime) and increasing future production through optimized well placement while drilling.
Recently, the petroleum industry has capitalized on remarkable information and collaboration technologies that bring vast amount of real-time data to the decision-makers. The goal is to improve their decision-making processes by integrating technology, people, work processes, and organizations. However, complexity of drilling operations and investment costs put industry in quest for a feasible method to valuate benefits. The prevailing evaluation methods are usually restricted to qualitative approaches, and thus are inefficient as decision aids. The paper presents a decision analytic approach to quantitative valuation of integrated operations.The proposed valuation method is illustrated by a case from improving decisions' quality in drilling operations in the North Sea. Improving the quality of operational decisions requires the drilling team to have access to the relevant information in a timely manner. The relevant information needs to be filtered, aggregated, and fused into decisions; and correspondingly represented to the decision-maker. Availability of new information allows making new decisions, and changing work processes. Adoption of a new tool, conformance to a new work process and trust in aggregated and shared information are typically critical success factors. The method is designed to qualitatively analyze changes in intangible capital and assess technology adoption under conditions of operational and economical uncertainties.The method accounts for information quality improvements and related uncertainties management. In turn, decision analytical part of the method provides a means to relate observable improvements in Key Performance Indicators (KPIs) and uncertainties. Operational decisions are formalized using influence diagram, a.k.a. Bayesian decision network.Implementation of digital fields and integrated operations is technologically and organizationally challenging. The functional steps of the established methodology and the decision analytic approach are significant contributions to understanding the quantitative value of these projects.
A systematic decision analytic methodology for providing insight into horizontal well placement decisions is presented. These decisions require access to real-time data and to input from subject-experts on the team and from the decision-maker. The available information and outcomes of alternatives are associated with uncertainty. Furthermore, the decision-maker may have multiple objectives, such as future productions, wellbore configuration, and drilling costs. The methodology uses Bayesian decision networks, also known as influence diagrams, to frame, analyze, and support operational decisions.Influence diagrams are compact graphical representations of decisions and are based on decision and probability theories. They rigorously illustrate the interrelationships among decisions, key uncertainties, and value functions. Furthermore, they allow incorporating sensor readings, sub-surface model outputs, and experts' knowledge, which enables probabilistic inference and decision support in real-time contexts. Finally, the influence diagrams models may be used to calculate value of information and of control.In this paper, we focus on supporting well placement decisions involving multiple objectives. We explore the feasibility and suitability of using two potential extensions of the influence diagrams: (1) multiple-attribute utility influence diagram and (2) multiple-objective influence diagram. Using the methodology, the example shows that both extensions have the potential to improve the decision-making process within the multi-disciplinary team. They can provide further insight into decisions by ranking the value of information for key uncertainties from the perspective of different sub-groups of the drilling team.Following a systematic decision analytic methodology, in particular, decision analysis cycle, and applying influence diagrams in drilling operations would create value by reducing losses (materials and drilling/completion processes' downtime) and increasing future production through optimized well placement while drilling.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.