SAS-Authored Papers for the Society of Petroleum Engineers
Oilfield Data Mining Workflows for Robust Reservoir Characterization: Part 2
Keith R. Holdaway, SAS
Abstract
To master complex problems inherent in heterogeneous subsurface reservoir systems, we must break down the walls built around the traditional disciplines of petroleum engineering, geophysics, geology, petrophysics and geochemistry. Professional curiosity is trumped by necessity to uncover the knowledge hidden across all upstream data sets as a multi-disciplinary analysis, underpinned by a multivariate suite of advanced analytical workflows, is implemented through data mining methodologies.
This paper sets out to answer some important questions around data mining workflows underpinned by exploratory data analysis, confirmatory data analysis, descriptive and predictive modeling to establish sound and important reservoir characterization decision-cycles. Case studies are presented to illustrate effective and successful studies based on advanced statistical analysis and AI workflows in sandstone and carbonate reservoirs.
Maximize the Placement of Wells and Production in Unconventional Reservoirs: Part 1
Keith R. Holdaway
Abstract
This paper walks through two case studies implemented in the Bakken and Pinedale assets in the United States, exemplifying data mining workflows that successfully improved hydrocarbon production.
It is critical to assess via data mining methodologies the variability in and potential of well performance in order to formulate an optimized suite of well completion and reservoir development strategies. Owing to the inherent complexity of subsurface systems, a data driven set of advanced analytical workflows that embrace exploratory data analysis in a multivariate perspective and predictive data analysis must be implemented in order to complement the first principals that underpin the array of geoscientific schools of thought.
Maximize Placement of Wells and Production in Unconventional Reservoirs: Part 2
Keith R. Holdaway, SAS
Abstract
This paper walks through two case studies implemented in the Bakken and Pinedale assets in the United States, exemplifying data mining workflows that successfully improved hydrocarbon production.
It is critical to assess via data mining methodologies the variability in and potential of well performance in order to formulate an optimized suite of well completion and reservoir development strategies. Owing to the inherent complexity of subsurface systems, a data driven set of advanced analytical workflows that embrace exploratory data analysis in a multivariate perspective and predictive data analysis must be implemented in order to complement the first principals that underpin the array of geoscientific schools of thought.
Predictive Analytics: Development and Deployment of Upstream Data Driven Models
Keith R. Holdaway, SAS
Abstract
Building data driven models that can predict future production under uncertainty conditions is essential to shorten critical decision-making cycles. With improved workflows and advances in High Performance Computing, it is now possible to ascertain risk and quantify uncertainty for very large populations of data without sampling and losing knowledge garnered by predictive models driven by the data and not by empirical petroleum engineering algorithms or deterministic methodologies.
This paper draws upon two case studies that ameliorate the path from raw data to invaluable knowledge. A suite of predictive models driven by real-time data have been implemented to identify optimized drilling and production strategies in the North American tight gas plays and acid stimulation strategies in the Gulf of Mexico.
This paper walks through two case studies implemented in the Bakken and Pinedale assets in the United States, exemplifying data mining workflows that successfully improved hydrocarbon production.
It is critical to assess via data mining methodologies the variability in and potential of well performance in order to formulate an optimized suite of well completion and reservoir development strategies. Owing to the inherent complexity of subsurface systems, a data driven set of advanced analytical workflows that embrace exploratory data analysis in a multivariate perspective and predictive data analysis must be implemented in order to complement the first principals that underpin the array of geoscientific schools of thought.
Data Mining For Safety
Bill Tuzin, SAS
Abstract
Virtually every company involved in the exploration and refinement of hydrocarbons has installed procedures and systems to measure workplace hazards and incidents. Monitoring incidents and hazards and performing timely analysis leads to effective remediation efforts. Many of these hazard identification systems and procedures are ineffective because they may be paper based, may be cumbersome for part-time computer users, and are often unavailable at the job site.
This paper identifies new technologies that have been implemented in other industries that could improve the performance and outcomes of safety systems used in the energy industry. The principle benefit is the ability to overcome the many obstacles that prevent identification and reduction of risk. By enabling easier data collection from virtually anywhere, with assurances that automated processes will consistently and correctly interpret the data collected, the outcome will be a process of prompt and accurate risk mitigation.
Automating Well Performance Monitoring of Real Time Data
Co-authored by Keith R. Holdaway, SAS, and Ahmed Al Nuaim, Mike Williamson, Sualeh Hasan, Marwan M. Labban
Abstract
Estimating reserves and predicting production in reservoirs has long been a challenge. The importance of performing accurate analysis and interpretation of reservoir behavior using only rate and pressure data as a function of time or cumulative production is fundamental to assessing remaining reserves and for forecasting production. An interactive system was developed to assist in Decline Curve Analysis (Cartesian, Exponential, Harmonic and Hyperbolic) and Production Forecast in an Integrated Reservoir Management environment. This paper introduces a refreshingly intuitive navigation of the disparate data sources and input parameters aligned with a comprehensive suite of forecasting models wrapped into a web-based solution.
Automating Decline Curve Analysis in an Integrated Reservoir Management Portal
Co-authored by Keith R. Holdaway, SAS, and Ahmed Nuaim, Husameddin Madani and Dennis Seemann, Saudi Aramco
Abstract
Estimating reserves and predicting production in reservoirs has long been a challenge. The ability to perform accurate analysis and interpretation of reservoir behavior using only rate and pressure data as a function of time or cumulative production is fundamental to assessing remaining reserves and forecasting production.
This paper introduces an interactive system designed to assist in decline curve analysis and production forecasting in an integrated reservoir management environment. It also discusses interactive graphical visualizations and extensive what-if combinations of critical input parameters across multiple forecasting models.
Drilling Optimization in Unconventional and Tight Gas Fields: An Innovative Approach
Keith R. Holdaway, SAS
Abstract
The majority of the technically recoverable natural gas is present in unconventional reservoirs such as tight sands, shale and coal beds. There are high costs associated with these unconventional and deep gas assets. To minimize drilling and completion times, constrain risks, quantify uncertainty and ultimately maximize the lifetime of a well's productivity, it is crucial to garner as much knowledge as possible from disparate and limited data sources.
This paper illustrates an advanced analytical methodology that aggregates and integrates data from a multitude of sources, performs robust quality control, and then correlates the information to discover patterns so that engineers can devise more efficient, accurate exploitation strategies. It proposes using an advanced analytical framework that introduces an exploratory data analysis step for reservoir characterization followed by determination of key production indicators and multivariate methodologies to classify wells. With improved understanding of inherent correlations between geology, reservoir, rock mechanics, the frackpack process, and proppant fluid and well performance, it is feasible to cluster wells in accordance with the overriding production indicators, thus dividing the field into clearly defined segments.
Enhance Digital Oil Fields by Plugging the Technological Capability Gap
Keith R. Holdaway, SAS
Abstract
The digital oil field is a strategy for improving a specific area of an oil company's business by effectively deploying people, technology and knowledge. A key ingredient of the digital oil field is quick, easy access to quality data.
To achieve the digital oil field, the oil and gas industry needs to plug the capability gap with pertinent technology that embraces current trends. Solutions must integrate not only data, but also extant applications and physical monitoring systems that can effectively educate and equip almost every aspect of upstream oil and gas activity into an integrated, real-time operation. This approach requires domain experts and technology to surface one collaborative perspective on a plethora of accumulated data – both real-time and historical. As a result, digital oil fields will engage both the geosciences community and business decision makers.
This paper highlights a case study about a reservoir surveillance management solution that embraces predictive detection and alerting to improve production forecasting and planning and to assist in managing corporate risks and investments.
Exploratory Data Analysis in Reservoir Characterization Projects
Keith R. Holdaway, SAS
Abstract
Characterizing the reservoirs of a mature field involves analyzing large data sets collated from well tests along with production history and core analysis results enhanced by high resolution mapping of seismic attributes to reservoir properties. Exploratory data analysis (EDA) is an overture to spatial analysis, simulation and uncertainty quantification that ensures consistent data integration, data aggregation and data management – underpinned by univariate, bivariate and multivariate analysis. As an approach to analyzing data for the purpose of formulating hypotheses that are worth testing, EDA complements the tools of conventional statistics.
This paper details some of the more common EDA steps that initiate efficient reservoir characterization projects. It also underlines the importance of the EDA school of thought, which is often overlooked or even precluded prior to the spatial analysis, kriging, simulation and uncertainty quantification steps.
Increased Upstream Asset NPV with Forecasting, Prediction and Operational Plan Adaptation in Real Time
Horia Orenstein, SAS
Abstract
Managing an operating asset in the oil and gas industry is a complex activity. The physical process needs to be controlled so that technical objectives can be achieved while a multitude of process parameters are kept within expectations. At the same time, all operating modes, as well as the transitions from one operating mode to another, have to comply with technical process feasibility, equipment reliability and integrity criteria; finally, the safety of the staff working on the asset, and the safety of the surrounding natural and social environment, must be monitored and controlled.
Because management of an operating asset is the joint effort of many disciplines and hierarchies, it requires data; coordination of people and their knowledge; procedures embedded in work and business processes; handling of information and communication technologies and related technical means; and interaction with the asset's equipment. This paper presents a practical case where an innovative asset operation optimization solution facilitated efficient, real-time decision making in planning the right actions for the right situations and in proactively preventing disruptions by predicting early performance deviations.
Attitude of Collaboration, Real-Time Decision Making in Operated Asset Management
Horia Orenstein, SAS
Abstract
To effectively manage an operated asset, it's necessary to coordinate and collaborate with all disciplines and hierarchies involved. One should also consider that in today's world, more operating modes are being added by market, political and social demand; the transitions from one operating mode to another occur more frequently; and regulations are growing tougher. This paper asserts that managers of operated oil and gas assets need to quickly manage disruptive events and potential opportunities by opportunistically adapting their plans for target achievements. A collaborative, real-time event management and plan adaptation is defined in this paper as managed collaboration and real-time decision making.
