Unlike most businesses right now, oil companies are hiring. They can't afford not to. Driven by a steady increase in demand, the energy sector is one of the few areas of the US economy that is in a growth cycle -- in fact, payrolls in the oil and gas industries have swelled by 10% since the beginning of the recession.

Growth is so dramatic that the industry can't keep up. At a recent trade conference, 72% of executives surveyed felt that the industry has not done enough to avoid a shortage of technical talent. It's especially worrisome considering that an estimated 50% of the industry's engineers are expected to retire over the next 10 years.

It's not a minor obstacle, nor are the problems associated with rapid growth. As large corporations add employees, workforce management becomes increasingly important. Given that there are so many unknown variables in the oil and gas industry -- economic events, natural disasters, and so on -- business intelligence is one of the most efficient ways to ensure that the workforce is effectively utilized.

John E. Kelly III, IBM senior vice president and director of IBM Research introduces IBM's Watson computer system to the media during a press conference at IBM's Watson Research Center in Yorktown Heights, NY, Jan. 13, 2011. Watson competed against Jeopardy!'s two most successful contestants. (Photo courtesy of IBM)

Watson

IBM hopes to revolutionize the availability and analysis of information through a new question-answering machine called Watson. Jeopardy! viewers may recall that Watson, the IBM computing system, beat record holders Ken Jennings and Brad Rutter in a three-episode contest broadcast in February 2011. For regular watchers of the show, it may have just been good entertainment. For anyone with any interest in the evolution of technology, it was an impressive feat.

One of many ways that Watson distinguishes itself from any old computing system is its ability to understand natural language. It can memorize immense amounts of data from a variety of unrelated sources and make sense of it. Similarly, Jeopardy! contestants must be able to access decades' worth of acquired knowledge. They must also have an in-depth understanding of context and language, and they need mental agility to answer questions in seconds.

Watson studied for the challenge by memorizing roughly 200 million pages of content that came from a range of sources, including screenplays and encyclopedias. On questions in which it was not entirely certain it had the correct answer, it took longer to hit the buzzer. When it was confident in its answer, it could hit the buzzer in as little as ten milliseconds. Equally impressive, Watson's accuracy rate -- while not perfect -- easily bests the vast majority of humans. And like (most) humans, Watson can learn from its mistakes.

Using Watson in oil and gas

Jeopardy! is obviously very different from the oil and gas business, but the ability to retain

or store accurate information and to analyze and access it is a common need for both game show contestants and energy executives. While most businesses should have some sort of information technology system in place, in many cases the data is trapped in outdated databases or it is too difficult to locate when it is needed. Sometimes executives have to navigate multiple systems to find relevant data; other times the pertinent information can't be found or retrieved given the overwhelming amount of data that has been collected.

Because the industry's boom and bust cycles are highly dependent on the economic environment, Watson could prove especially useful in anticipating hiring needs at any given time. It can take relevant economic data from both internal and external sources, such as airlines' monthly traffic reports or internal sales reports, and try to project future oil demand. It can help guide hiring practices, and it can also draw from historical data -- month-over-month, year-over-year, or decade-over-decade -- to answer complex business questions. Equally important, because Watson understands natural language, users can pose questions as they would to a peer (e.g. "How many drilling engineers will I need to hire in three months?"); it doesn't require employees to learn yet another database language to use it.

Watson is also poised to dramatically improve employees' safety and productivity while reducing costs. As oil companies turn to increasingly hostile environments to explore and drill, Watson can offer detailed guidance on how to operate in different locations based on data collected about working conditions. Knowledge that is often kept in a closed system for specialists of specific geophysical disciplines could be accessed by Watson and therefore available for employees in other areas to use. As it stands now, the average oil engineer spends roughly 60% of his time on data mining; that's time that could be spent doing far more valuable work. In the event of a crisis, Watson can look at similar historic events and suggest the most effective courses of action.

Dr. Eric Brown from IBM Research preps four members of congress -- Jared Polis (D-Colo.), Bill Cassidy (R-La.), Jim Himes (D-Conn.), and Rush Holt (D-N.J.) -- for an exhibition game against IBM's Watson Feb. 28, 2011. The match fostered a conversation among government leaders about the importance of IT to US global competitiveness and encouraged greater focus on math and science education. Final score: Watson $40,300, Congressional Members $30,000. (Feature Photo Service)

The question-answering machine can also assist in age-old workforce challenges such as scheduling -- if there's a shortage of employees, how does a company maximize productivity with current workers? If there are operational problems in one area, how can idled employees be utilized while waiting for issues to be resolved? Complex questions that might have taken several employees several hours to figure out in the past can now be calculated by Watson in seconds.

Just as IBM's seismic modeling technology has evolved, there's reason to expect Watson will, too. It's not outlandish to assume that in the near future Watson will recognize voices and understand spoken English as well as other languages. Managers out on the field may be able to use handheld devices to phone questions in, tapping a wealth of knowledge that was previously unavailable to them.

The oil industry may face complex problems, but workforce management should not be one of them. Watson can help executives allocate resources while reducing costs and increasing productivity. Plus he's great with trivia.