BP is putting “big data” at the centre of its push to lower costs, as it tries to harness digital technology to lift efficiency in an era of weak oil prices.
The UK oil group is planning a fivefold increase in its data capacity over the next three years, and Ahmed Hashmi, head of technology for BP’s exploration and production business, claims the biggest efficiency gains are still to come.
BP is aiming to raise its data storage capacity from about 1 petabyte — equivalent to 20m four-drawer filing cabinets filled with documents — to 6PB by 2020 as the group integrates machine learning and artificial intelligence into its operations.
More than 99 per cent of oil and gas wells operated by BP around the world are equipped with sensors that produce a constant flood of real-time data on production performance as well as the condition of infrastructure, Mr Hashmi said. This information is fed into a cloud-based storage system which allows BP engineers anywhere in the world to access the information.
“The vision is to have absolute knowledge of what’s going on in the field,” Mr Hashmi told the Financial Times. “Increasingly, we can improve reliability and safety by predicting what is going to happen before something goes wrong.”
All large oil and gas companies are investing in digital technology as part of wider efforts to increase efficiency at a time of weak crude prices and rising competition from low-cost US shale resources and fast-growing renewable energy.
However, Mr Hashmi claims BP is ahead of most rivals after spending “several hundred million dollars” equipping its facilities with sensing devices and 2,000km of fibreoptic cable carrying 5m data points every minute.
As well as commercial incentives, BP has faced added pressure to improve monitoring after the blowout on its Deepwater Horizon platform in 2010 killed 11 men, spilled 3m barrels of oil into the Gulf of Mexico and cost the group $62bn in fines, compensation and clean-up costs.
Reliability of BP’s exploration and production facilities — measuring factors such as time lost to production stoppages — has increased from 88 per cent in 2012 to 95 per cent last year and Mr Hashmi said technology was the main reason. This had contributed to $7bn of annual cost savings by BP since 2014.
Among the innovations has been a “digital twin” system that allows BP engineers to test maintenance procedures and other critical engineering work using virtual reality before carrying out the work on real facilities.
In June, BP invested $20m in a Californian start-up called Beyond Limits, which is aiming to commercialise artificial intelligence technology developed for deep space missions by Nasa. Mr Hashmi said machine learning would allow BP to use accumulated data from drilling operations to improve speed and success rates in future wells.
Automation would reduce the number of people needed to operate some facilities, including offshore production platforms, Mr Hashmi acknowledged, but he insisted this was not the main objective.
“We’re putting data at the fingertips of our engineers and scientists,” he said. “We are allowing them to spend more time doing high-value work rather than putting Excel spreadsheets together. We see a symbiotic relationship between machines and humans where artificial intelligence optimises the choices for people to make.”
Share this:
