BP is using Azure machine learning (ML) to reduce the time needed to select precise prediction models and boost the productivity of its data scientists. BP’s principal for digital innovation, Manish Naik, spoke to Microsoft’s Bill Briggs about the benefits of using machine learning for ‘drilling down into its data’.
In order to extract gas and oil to light cities, transport people and run industries, engineers like those at BP must locate the reservoirs and accurately predict what percentage of hydrocarbons are retrievable – “recovery factor”. This task has traditionally been iterative, resource-heavy and can often involve human bias. Data scientists can spend weeks trying to find the best prediction models. Naik explains that using Microsoft Azure ML can reduce this time from weeks to day and days to hours, as well as removing the human bias.
The model that BP developed for recovery factor is now “in production and used daily by hundreds of subject matter experts at BP from all around the world,” said Naik. “It will make data scientists more productive, which means faster time to market for ML projects.
“As data scientists continue to use more automated machine learning, they will develop trust in the output it provides. In the future, this will form a part of a robust benchmarking process for all ML projects, thus improving quality.”
The oil and gas industry generates a huge amount of data and Naik explains that there are therefore “lots of opportunities to exploit this data using AI, ML and cloud technologies,” he said. “In broad terms, there is significant potential for these technologies to help improve the efficiency of our operations and help us make better, more accurate and informed decisions.”