SPE logo

Next Generation of Smart Reservoir Management:
The Eminent Role of Big Data Analytics

18–21 January 2016 :: Dubai, UAE

Technical Agenda

Monday, 18 January 2016, 0830–0900

Keynote Session

Khalid Al-Subai, Saudi Aramco

Satyam Priyadarshy, Halliburton

Monday, 18 January 2016, 0900–1230              

Session 1: Data Quality, Validation, and Standardisation

Session Manager: Ahmed Hutheli, Saudi Aramco

With the implementation of intelligent fields and state-of-the-art technologies for best-in-class reservoir management practices, data quality assurance and standardisation become imperative.

This opening session examines the key factors for establishing a robust and reliable data management system that has the resilience and scalability to handle and deliver consistent data. We believe such a system should have the following factors: data standardisation, data quality control, data filtration, data access and visualisation, and system reliability and availability. In the context of each factor, associated challenges will be identified and assessed. Moreover, the session will discuss how the actual impact of each factor on data quality assurance depends on the conditions of any particular case. For instance real-time downhole sensor capture undergoes through data quality assurance processes that are different than manual data capture processes. Hence, only a subset of the mentioned factors may be relevant in that scenario. Therefore, it is important to understand the risks to data quality in each specific case. Implementing and understanding such system will provide reliable and consistent data which will result in fast and optimised decision making and ultimately will contribute significantly in increasing production potential, recovery factor, and efficiency.

1400–1730

Session 2: Integration of Disparate Data Sources regardless of Origin, Time Scale, or Structure

Session Managers: Ady Al-Sharif, Maersk Oil; Morgan Eldred, Gartner

Every well planned, drilled, completed, put into production and later abandoned, generates a huge amount of data at every stage of the process. Upscale to the field level and terabytes of data is available to engineers and scientists, and it seems likely that the future of field development will continue to drive the collection of ever more varied data—especially with the growth of automation and advanced analytics.

Increasingly, such operations are characterised not only by the size, but also the variety of the associated datasets: not all data is in structured rows and columns, but in unstructured images, video, text, work logs, streams and more. The full value of these disparate datasets can be significant, but is only realised when data is integrated and seen in context…in context of each other and in context of relevant time, role-centric workflows. The opposite is also true: unless usable information is produced, data collection can be seen as a burden and waste of time and company budgets.

The discussion in this session will focus on the techniques and processes that business professionals will need to develop and employ to properly capture and transform complex, multi-timescale, structured and unstructured, and other highly variable data into reliable information that can be used to create insight in taking profitable actions.

Tuesday, 19 January 2016, 0900–1230

Session 3: Data Mining, Statistics, and Predictive Modelling

Session Managers: Detlef Hohl, Shell; Marko Maucec, Saudi Aramco

Recently, the pursuit of excellence in the upstream petroleum area has been reinforced by data-driven sciences such as descriptive statistics, automatic-process-control theory, data mining, and predictive-analytics-modelling techniques. The development and production of oil and gas resources supported by continuous real-time data entail learning and knowledge search over a relatively long period of time. As we drill more wells and produce for various decades we acquire more data and information, progressively reduce the uncertainty in production forecasts and continuously improve the economic value of oil and gas fields. However, challenges persist in identifying and characterising the basic modelling elements affecting petroleum production forecasts in a multidisciplinary fashion.

This session will explore the cross-link between fundamental sciences and rapidly changing digital-information technologies in the area of hybrid approaches to oilfield-knowledge management by addressing questions like:

  • How could data analysis refine physical models and determine values or related parameters so robust that reliable predictions can be made?
  • How data are validated and interpreted and how results are communicated fast enough to make the right decisions at the right time in the right context?
  • How can we more efficiently analyse existing and new data using cross-validation approaches and dimensionality-reduction modelling to uncover patterns, useful for predictions on production and recovery?
  • How can we adopt and implement best-practice technologies from fellow industries and crate a game-changing momentum for the oil and gas business?

1400–1730

Session 4: Scalable Big Data Analytics for Production and Operational Excellence

Session Managers: Stan Cullick, Greenway Energy Transformations; Sebastien Matringe, Quantum Reservoir Impact

The session will focus on production: surveillance, diagnostics, optimisation and operations, failure tracking and analysis, systems and equipment reliability, EH&S, and field execution.

Real-time production instrumentation, i.e. pressure, multi-phase flow, temperature, vibration, erosion, pump performance, power, etc., from the entire value chain of completions, well heads, flow lines, batteries, separators, etc., is loading databases with terabytes of data.

In addition to this real-time data, massive operational historic data, operational engineering models, digital and non-digital context unstructured data and event data from across the well/field life-cycle must be integrated for full utilisation with this high density digital data.

The analytics tools of the future must be able to scale the data for production decisions, and engineers must be able to fully analyse all of these data elements with highly advanced analytics and models to better understand correlations and inter-dependencies that will enable a more proactive, predictive, and prescriptive operations environment.

The future of operations will require advanced analytics for equipment tracking and reliability statistics and predictive analytics to improve operational efficiencies. Operations will be advancing to use of extensive automation, closed loop control and even robotics to operate equipment, service wells, etc. which will require new ways of looking at and using the data and analytics in a highly proactive and predictive manner. This session will discuss technologies to advance production and operations as described above.

Wednesday, 20 January 2016, 0900–1230

Session 5: Large-Scale Data Analytics for Portfolio Management

Session Managers: IBM; Teradata

As competition for scarce resources intensifies, portfolio management is becoming one of the most critical analytical exercises in our industry. For some companies, a strategic shift from an integrated business model to a pure-play E&P operations model gives sharper focus to the imperative of developing high-performing assets, while uncertainty persists in how to optimise their portfolio from harder to reach and harder to manage assets in mature fields, offshore fields, and unconventional plays. 

More data and analytical modelling techniques are available to portfolio managers than ever before.  New technologies are delivering new measurement and modelling techniques,  the data growth trend is set to continue, and these managers are expected to utilize ALL of this data to make faster, better decisions on company portfolios and investments that can have business impact for years if not decades. This session will explore the techniques that analysts will need to master in order to:

  • Leverage advanced analytics and other methodologies to derive the best possible information from available data
  • Use that information in decision analysis regarding portfolio investment and which assets to acquire, develop, retain, or sell.

1300–1530

Session 6: Advances in Data Visualisation

Session Managers: Dave Stern, ExxonMobil; Stan Cullick, Greenway Energy Transformations

A critical challenge in applying advanced analytics to reservoir management is developing tools that can be used by practicing reservoir engineers and geoscientists to understand the results of analysis, and ultimately use those results to make decisions to optimise reservoir performance. Current 2D and 3D visualisations and dashboards may evolve to much higher dimensional interactive visualisations in the future. Key features of these tools are the ability to visualise disparate and dense, sometimes high frequency data, to easily qualify and to see trends in high-dimensional parameter space, find anomalous results, integrate disparate data types, and visualise predicted scenarios with associated uncertainties.

Discussion in Session 6 will explore avenues for development of new tools to enable more efficient and effective utilisation of big data and to provide visualisations of “what-if predictive modelling”.

.

Thursday, 21 January 2016, 0830–1200

Session 7: Impact of Access to All the Data All the Time to All the Stakeholders

Session Managers: David Holmes, EMC; Nasser Mahrooqi, Petroleum Development Oman

According to IDC, “The 3rd Platform is built on the technology pillars of mobile computing, cloud services, big data and analytics, and social networking. Adoption is being driven by business requirements for mobile access by a distributed workforce, enhanced collaboration, and predictive analytics to anticipate issues and prioritize decisions for resolution—made available for cloud-agnostic deployment to mitigate implementation complexity and risk.”

The objective of this session is to understand the 3rd Platform, its potential impact on our industry and this forum’s focus areas, and to discuss and develop ideas on how to harness the power and potential of the 3rd Platform to maximise results.

With the promise and the premise that all data will be available to all users and stakeholders in all of its forms (from raw to interpreted to real-time to correlated models to advanced analytics to beyond) to generate more proactive insight and make better decisions faster, how will our industry need to “change” to be able to truly utilise all the data in the context of:

  • Individual contributors, core workflows and activities
  • Cross functional, cross discipline teams, and assets
  • Improved collaboration and collaborative decision making
  • Real-time and relevant time predictive and prescriptive analytics
  • Advanced monitoring and control, including closed-loop control
  • Operational technology, data management, and user experience requirements to reach the potential “connected decisions” model
  • Improved business impact and bottom line results resulting from such approach

1400–1730

Session 8: Impact of Access to All the Data All the Time to All the Stakeholders

Session Managers: David Holmes, EMC; Nasser Mahrooqi, Petroleum Development Oman
 
According to IDC, “The 3rd Platform is built on the technology pillars of mobile computing, cloud services, big data and analytics, and social networking. Adoption is being driven by business requirements for mobile access by a distributed workforce, enhanced collaboration, and predictive analytics to anticipate issues and prioritize decisions for resolution—made available for cloud-agnostic deployment to mitigate implementation complexity and risk.”

The objective of this session is to understand the 3rd Platform, its potential impact on our industry and this forum’s focus areas, and to discuss and develop ideas on how to harness the power and potential of the 3rd Platform to maximise results.

With the promise and the premise that all data will be available to all users and stakeholders in all of its forms (from raw to interpreted to real-time to correlated models to advanced analytics to beyond) to generate more proactive insight and make better decisions faster, how will our industry need to “change” to be able to truly utilise all the data in the context of:

  • Individual contributors, core workflows and activities
  • Cross functional, cross discipline teams, and assets
  • Improved collaboration and collaborative decision making
  • Real-time and relevant time predictive and prescriptive analytics
  • Advanced monitoring and control, including closed-loop control
  • Operational technology, data management, and user experience requirements to reach the potential “connected decisions” model
  • Improved business impact and bottom line results resulting from such approach