machine learning and big data analytics

How Do Energy Companies Use Machine Learning? GE Canada Oil & Gas General Manager Shaun Kelly Explains

At the Customer Innovation Centre in Calgary, one of the coolest things we are developing is in the field of machine learning and advanced analytics, writes Shaun Kelly, General Manager at GE Oil & Gas Canada. 

Optimizing production—whether that be in a drilling campaign, SAGD, artificial lift, or in a refinery—requires qualified engineering talent to make calculated decisions. These decisions are based on engineering principles, but they also usually involve many variables and KPIs, and some of these KPIs directly conflict with each other. That is, a decision that might be best for one step of the process could have a negative impact somewhere else in the process.

So, the engineer uses his or her experience and knowledge to iterate to the best solution. This could take one person days or even weeks to get to the optimized state.

This is where machine learning comes in, because machines are much better than humans at iterating. Machine learning has advanced to the point where large quantities of structured and unstructured data can be ingested and synthesized, millions of hypotheses can be tested, and insights and outcomes derived in seconds.

With machine learning, we’re not wasting the time and talent of engineers on these types of tasks. Machine learning doesn’t replace humans; rather, it’s a tool that allows them to accomplish much more, faster. By optimizing production in this way, we’re boosting efficiency and saving money for the oil and gas industry. It’s a true game-changer.

Got questions? Just #askGE in the comments below, or find us on Facebook, Twitter, or LinkedIn. A team of GE experts will be answering your questions on Twitter on May 2nd.

Want more stories like this delivered to your inbox?

subscribe