From exploration to accounting, IT helps companies achieve goals

More than 80% of petroleum executives say ‘technology progress' will be the most important external force for the industry by 2030.

John D. Brantley, IBM, Southbury, CT

One of the greatest challenges in oil production is that it is wildly unpredictable. Market volatility, pricing volatility, natural disasters, man-made disasters, geopolitical conflicts, and ever-changing regulatory environments are common hurdles for energy companies.

While technology has made most aspects of the business far more efficient, there is much room for progress. By one estimate, only 33% of oil is pulled from existing reservoirs, which suggests recovery rates could be dramatically improved. Even a 1.5% increase in recovery rates could yield enough oil on average for a half year of global consumption. And we have reason to believe that information technology can help producers achieve these goals.

As demand for oil increases with every passing day, so does the need for greater efficiency. Current projections suggest demand for energy will grow by 50% from 2005 to 2030. Most executives not only hope technology can help meet the growing demand for energy, they expect it. Roughly 81% of oil industry executives polled by IBM said "technology progress" will be the most important external force for the industry in 2030. It's not just exploration that can be supported by technology – it's every aspect of the business, ranging from the refining process to accounting.

IBM has a long history with oil producers, going all the way back to 1912, when the Atlantic Refining Company (now British Petroleum) installed a tabulating system made by C-T-R, IBM's precursor company. In 1938, IBM was in Venezuela, operating under the name C.A. Watson de Maquinas Comerciales, when it sold Mene Grande Oil Company its first IBM machine.

As technology's role in oil exploration has taken on greater importance, IBM has kept up with the changing needs of energy companies. Case in point: Repsol used an IBM processor to discover oil reserves buried 30,000 feet below the Gulf of Mexico's surface in 2009. The reserves were found using an innovative and computing-intensive process known as reverse time migration (RTM), in which seismic data is translated into 3-D maps.

Even 15 years ago, Repsol might not have been able to make the discovery with the technology available at the time. Since all the "easy oil" has been found, producers are exploring increasingly hostile environments – both politically and geologically – which means dry holes are an increasingly expensive risk. Businesses do not have the leisure to wait around for a supercomputer to determine whether a potential location does or does not have oil.

As it stands now, it's not unusual for dry hole expenses to gobble up millions of dollars out of an energy company's net income – one dry hole can cost as much as $125 million. The losses aren't just financial, though – they consume time and resources, and then there are the losses related to missed opportunities.

The risks of dry holes can be minimized, though, thanks to the computing power available today. The amount of time it now takes to run complex imaging algorithms to find oil could be as little as two weeks, where it might have taken months to run the same algorithm only 10 years ago.

Because oil prices are the most important determinate of energy companies' profits, it's important for producers to curb expenses and grow revenue during periods of pricing volatility. Without the help of a crystal ball, trying to forecast oil prices is a difficult, if not impossible task. There may not be a means of predicting the future with 100% accuracy, but we can rely upon data analytics to give us a fairly decent idea of where things will go. Using historical data as well as current, real-time data, we can build pricing models using predictive software. Armed with these projections, executives can make informed decisions about how to balance cash flow with capital investments. In some cases, executives may find the cost of drilling at certain locations could exceed potential profits due to weak prices.

Watson, a question answering machine developed by IBM, is also poised to dramatically change the business of oil production. Fans of the TV show Jeopardy! may have seen Watson compete – and beat – record-holders Ken Jennings and Brad Rutter in February. The computer system, which has mastered natural language English, learned more than 200 million pages of content (including screenplays and encyclopedias), and sifted through all of its knowledge to answer questions in fractions of seconds.

Just as Watson absorbed vast amounts of information from a variety of origins to compete on Jeopardy!, it can take in data streams from disparate sources to help resolve problems with tremendous accuracy and speed. Users would be able to ask questions of it in the same way they would ask questions of a colleague, such as: "How many reservoir engineers will we need at X time?" Watson could help identify and manage well performance, take data from real-time readings (such as flow of oil, water and so on) and use predictive modeling techniques to suggest ways to improve recovery rates.

Energy companies already use information technology to help find oil and to manage drilling operations. The question is whether their data is accurate and well managed. Sometimes human error – such as incorrect input of any given number – can throw off operations and increase expenses.

Companies with high dry-hole rates either misinterpret their seismic data or their data is erroneous. Sometimes they hit dry holes because they don't dig deep enough, or they may be drilling in the wrong location. In any case, a computing system like Watson could reduce costs as well as the risks of dry holes and improve profitability.

Equally important, Watson may help provide specific information to employees on the field when they need it. If knowledge is isolated or locked up, it's useless, and given that oil production requires such a wide range of specialists, every single employee's expertise is likely of value to someone else in the organization.

In the case of Norwegian energy company Statoil, all of its offshore platforms operated independently so it was difficult to develop a set of standardized processes. IBM helped design a system that allows different units within the company to share data and collaborate. The upshot: Improved productivity and lower costs and safer operations. The system has the potential to generate $50 billion in value over the next five years.

Right now, Watson is only accessible via computer, but the next step is to develop voice recognition software which would allow users to verbally ask Watson questions just as they might ask questions of a co-worker – a very, very smart co-worker who could potentially save the company millions of dollars.

About the author

More Oil & Gas Financial Journal Current Issue Articles
More Oil & Gas Financial Journal Archives Issue Articles
View Oil and Gas Articles on PennEnergy.com