In an upstream investor presentation, which was sent to Rigzone by the BP team earlier this year, Gordon Birrell, BP’s EVP of Production and Operations, said the company was “busy developing the next wave of technologies” and touched upon a few artificial intelligence (AI) developments at BP.
In order to find out more about BP’s upstream AI capability, Rigzone sat down with Jack Cameron, Vice President at BP Technical Solutions India, for a Q&A session, delving into what AI technology the company is using, where it came from, and BP’s future AI technology plans.
Rigzone: What AI technology is BP using in its upstream division and how does this help the company?
Cameron: Some of our most impactful use of AI is allowing us to maximize oil and gas production by helping us fully optimize our wells and facilities.
Optimization Genie, for example, is an enhancement to our industry leading real-time digital hydraulic twin and helps us determine how our production operations can be adjusted to safely produce more oil and gas. It autonomously identifies and tests production system bottlenecks, simulates adjustments, and recommends optimal configurations for engineers to validate and implement. It’s fast; what used to take engineers weeks of manual analysis, Optimization Genie accomplishes in hours.
In the Gulf of America, on our Atlantis facility, it helped prioritize and optimize well variables at no extra cost that led to an increase of approximately 2,000 barrels per day.
Another example is Wells Assistant, which is like using Copilot or ChatGPT specifically for wells, giving users access to 100 years’ worth of institutional knowledge so they can get quick and accurate answers to nearly any query about wells, from the simple to the more complex.
With around 15,000 unique data points on file, it helps make our teams more efficient and preserves critical institutional knowledge.
Rigzone: When and how was this technology developed?
Cameron: Optimization Genie was developed in early 2024 by BP’s production and technology teams. It was piloted on the Atlantis field in the Gulf of America, showing its efficiency in identifying and simulating production enhancements. By September 2024, it was fully implemented and delivering gains. It was designed as an enhancement to BP’s real-time digital twin platform, APEX, which simulates oil and gas production systems.
Wells Assistant, also launched last year by BP’s digital technology team, supports AI integration across Wells and Subsurface and was developed using Frontier Large Language models and our proprietary data systems. It is connected to five major BP wells data systems and has answered more than 14,000 queries so far.
Wells Assistant and our dedicated subsurface AI assistants are available to more than 2,000 users, answering hundreds of questions daily and helping us embed best practices globally, build capability rapidly among our junior staff, and help make our operations more reliable.
Rigzone: Is BP ramping up its use of AI in upstream?
Cameron: Yes, we are. Between 2022 and 2024, we increased BP operated production by around four percent and protected around 10 percent more from going offline through surveillance and real-time monitoring and analysis.
We now have acoustic sensing installed on over 400 offshore wells to continuously monitor for sand incursion into the well. This allows us to proactively intervene and aim to avoid costly outages, which we believe to be unique across the industry.
We’re incorporating predictive analytics, using unstructured data to forecast equipment failures, improving reliability and encouraging early intervention.
We are embedding data scientists and engineers on-site, conducting bootcamps and hackathons to solve operational issues, and identifying safety risks.
Rigzone: What are the biggest challenges BP has faced when implementing AI in its upstream operations?
Cameron: Adopting new technologies such as AI requires our engineers to challenge themselves and commit to working differently and often this requires a shift in mindset. AI recommendations can differ from traditional engineering intuition, and we find that being problem focused and upskilling our teams can help create this culture change. It’s possible, but it takes time. Another challenge is ensuring the tool’s safety and reliability, a process that can take time.
Rigzone: Does BP plan to launch any other AI technology for its upstream division in the near future?
Cameron: Yes, we will continue to use AI responsibly in our operations. This will be aligned to specific needs of the business and where it is going to add the most value.
Cameron has been with BP for almost 20 years. He has previously managed operations across offshore platforms in the Gulf of America, led risk and integrity programs, and held additional roles at BP in operations and process engineering in the North Sea and Egypt.
To contact the author, email andreas.exarheas@rigzone.com
element
var scriptTag = document.createElement(‘script’);
scriptTag.src = url;
scriptTag.async = true;
scriptTag.onload = implementationCode;
scriptTag.onreadystatechange = implementationCode;
location.appendChild(scriptTag);
};
var div = document.getElementById(‘rigzonelogo’);
div.innerHTML += ” +
‘
‘ +
”;
var initJobSearch = function () {
//console.log(“call back”);
}
var addMetaPixel = function () {
if (-1 > -1 || -1 > -1) {
/*Meta Pixel Code*/
!function(f,b,e,v,n,t,s)
{if(f.fbq)return;n=f.fbq=function(){n.callMethod?
n.callMethod.apply(n,arguments):n.queue.push(arguments)};
if(!f._fbq)f._fbq=n;n.push=n;n.loaded=!0;n.version=’2.0′;
n.queue=[];t=b.createElement(e);t.async=!0;
t.src=v;s=b.getElementsByTagName(e)[0];
s.parentNode.insertBefore(t,s)}(window, document,’script’,
‘https://connect.facebook.net/en_US/fbevents.js’);
fbq(‘init’, ‘1517407191885185’);
fbq(‘track’, ‘PageView’);
/*End Meta Pixel Code*/
} else if (0 > -1 && 63 > -1)
{
/*Meta Pixel Code*/
!function(f,b,e,v,n,t,s)
{if(f.fbq)return;n=f.fbq=function(){n.callMethod?
n.callMethod.apply(n,arguments):n.queue.push(arguments)};
if(!f._fbq)f._fbq=n;n.push=n;n.loaded=!0;n.version=’2.0′;
n.queue=[];t=b.createElement(e);t.async=!0;
t.src=v;s=b.getElementsByTagName(e)[0];
s.parentNode.insertBefore(t,s)}(window, document,’script’,
‘https://connect.facebook.net/en_US/fbevents.js’);
fbq(‘init’, ‘1517407191885185’);
fbq(‘track’, ‘PageView’);
/*End Meta Pixel Code*/
}
}
// function gtmFunctionForLayout()
// {
//loadJS(“https://www.googletagmanager.com/gtag/js?id=G-K6ZDLWV6VX”, initJobSearch, document.body);
//}
// window.onload = (e => {
// setTimeout(
// function () {
// document.addEventListener(“DOMContentLoaded”, function () {
// // Select all anchor elements with class ‘ui-tabs-anchor’
// const anchors = document.querySelectorAll(‘a .ui-tabs-anchor’);
// // Loop through each anchor and remove the role attribute if it is set to “presentation”
// anchors.forEach(anchor => {
// if (anchor.getAttribute(‘role’) === ‘presentation’) {
// anchor.removeAttribute(‘role’);
// }
// });
// });
// }
// , 200);
//});
