Chevron: Oil, Gas Cos Should Date, Not Marry, Big Data Tech Vendors

With 1,200 vendors available, oil and gas companies should date, not marry, their Big Data technology vendors.

Like other oil and gas companies, Chevron is swimming in data that represents a big opportunity for the industry, Irina Prestwood, analytics team lead for Chevron, told attendees at the second annual Data Driven Production Conference in Houston Wednesday. Prestwood quoted a Manhattan Institute report, Shale 2.0: Technology and the Coming Big Data Revolution in America’s Shale Oil Fields, which outlines the opportunities that industry has to tap shale resources using Big Data tools.

With data not just coming from sensors anymore, many companies are starting to recognize that Big Data tools could be a ‘Kodak moment’ for some, Prestwood commented. But the oil and gas industry must get a better handle on continuous monitoring and alarms, the ability to remotely open and close equipment, and optimize maintenance and diagnostic repair.

Full autonomy is where Chevron would like to go.  By making small changes, cost savings can be achieved. One example is more effective negotiation for piping and compressor equipment. Another area where Big Data can make a difference is human behavior. By evaluating successful projects versus unsuccessful ones, it’s most to determine if any unnecessary procedures can be safely eliminated. Analyzing wells also can help companies identify which wells need workovers.

Prestwood shared Chevron’s technology strategy for Big Data to achieve cost savings and enhance revenue. The first part of the strategy is to have a common data science platform. Second, Chevron needs to get a good handle on structured data. Third, the company needs to hurry up and start addressing streaming data. The fourth component is having the right people for the job; the fifth is having a plan to support this goal.

A common data science platform allows for a common set of tools to capture, store and perform analytics on structured and unstructured data. The common platform allows multiple customers access t to data with very little upfront effort.

With so many vendors to choose from, companies should do their homework on vendors. A common platform allows vendors to be changed with minimal impact, Prestwood explained.

Defining governance is another crucial part of Big Data strategy. A lack of definition will cripple a company’s ability to scale and size data. Other things to consider are whether all data should be loaded into a data lake, how many data lakes a company should have, how long data should be kept, who owns the data and how it’s made available to users.

Whatever platform or amazing analytics a company uses, if it’s not presented in a clear way, it may degrade the value of that data, Prestwood said. Chevron is quickly finding out that any analytic problem involving streaming and unstructured data requires all types of techniques to process and analyze this data.

What is the biggest bottleneck? Organizational capability, said Prestwood, who cited a 2012 McKinsey & Co report that the United States alone faces 50 to 60 percent gap between supply and demand of data scientists. To address the issue, Chevron conducts yearly data science competitions in which anybody within the company can take part. The company also has a comprehensive data science development program. Through this six-month program, experienced data science mentors from all areas and departments are paired with mentees. In some cases, people have switched careers after going through the program. Projects that have delivered value for the company also came out of this program, Prestwood said. Chevron also is training its data scientists to work with 21st century tools on the common platform.



Generated by readers, the comments included herein do not reflect the views and opinions of Rigzone. All comments are subject to editorial review. Off-topic, inappropriate or insulting comments will be removed.