Dr. Duncan Irving and Aurelien Guichard, Teradata’s Oil & Gas Industry Consultants, look at analytics for E&P operators
The value of big data analytics is quickly realised within the oil & gas industry, where operations are fast moving, complex and large in scale, often requiring massive engineering undertaking.
Today, major oil companies such as ExxonMobil, Royal Dutch Shell, Chevron, BP Total, ConocoPhillips, are investing in technologies that can help in the search for new reserves and also enable them to operate faster, more efficiently, and crucially, with lower risk and greater safety.
Hardware and software spending is and has always been massive in E&P.
For instance, in December 2012, BP announced that it will spend $100 million in its quest for oil using ‘the world’s largest supercomputer for commercial research’.
Armed with the new equipment, BP estimates that it will be able to perform 2,000 trillion calculations a second, reducing the time taken to perform seismic imaging of subsea rock formations from a few years to a few days.
We have seen large operators combine analytics technologies with sophisticated front-end dashboards and simulation models to determine optimal times and even work plans for well work-overs, water flood injection schemes, or critical equipment replacement, before it fails.
In the near future, analytics will yield substantial rewards, helping operators to make decisions relating to exploration, logistics and asset management, and also to generate important information relating to possible future outcomes.
This could include, for instance, enough detailed information to answer highly-specific and important questions such as: Does the platform need to be shut down to carry out this particular activity? What will be the effect for reservoir management? And if production is altered, what is the most favorable scenario with respect to safety, operating conditions and quarterly production revenue forecast?
During the planning stage for wells, the intelligent application of analytics offers the real potential to save vast amounts of time and cost by enabling more effective extraction from fewer and better planned wells. This also presents significant environmental advantages.
Article continues on next page …
In the drilling phase, the application of real-time analytics permits constant monitoring of down hole conditions based on constant data mining of all other drilling data for similar patterns. This, in turn, offers insight to the drilling campaign even as the drill bit is descending.
Other sophisticated capabilities of real time big data analytics include establishing the location of faults in pipeline infrastructure, based on understanding emerging patterns in sensor networks along the pipeline, serving to avoid costly downtime or, even worse still, environmental damage.
Similar approaches have already been successfully applied in other industries such as the
logistics, manufacturing and utilities sectors in order to spot emergent behaviour that may require rapid decision-making.
The growing requirement for data-driven decisions to be made in hours or days, rather than weeks or months, has served as a big motivating factor for the oil & gas industry to try to overcome these challenges and establish large-scale, highly operationalised analytics capabilities.
In a competitive scenario, the ability to appraise an asset more rapidly in an efficient, accurate way with uncertainties is vital.
In the licensing process, for example, a detailed evaluation of the economic potential and gaps in knowledge enable an asset portfolio to be managed in part using analytics, thus enabling an operator to divest or grow assets in line with market conditions. Historically there have been multiple barriers to leveraging analytics capabilities in the oil and gas industry, many of which remain true.
First, the sheer volumes of generated imaging and production sensor data can total tens of terabytes in a single imaging file or single daily stream, making the information extremely difficult to process. In addition, the data is rarely properly integrated, an essential requirement for meaningful analytics.
Second, the fact that collaboration must occur between people in a huge variety of different industry disciplines is also a massive obstacle as, for instance, geophysicists and engineers operate in very different ways and use their own distinct terminologies.
This, in turn, results in highly compartmentalised and specialised sets of industry data and applications, further exacerbating the big data analytics challenge.
In light of these seeming barriers to technology adoption, the excitement and awareness around the use of analytics continues to grow stronger as organisations realise the substantial value that these technologies bring to their ability to expedite intelligent decision making.
Operators that continue to develop their technical ecosystems with analytical practices improve their competitive advantage by allowing analytics across compartments. These organisations reap tangible benefits of driving faster, and more informed decisions, while maximising their ability to exploit knowledge.
