Two years ago, through demand generated within its business teams, multinational oil and gas supermajor bp recognized AI as one of its top strategic priorities. The operating model developed to accelerate this initiative is called ai@bp and comprises a core team that looks holistically at the opportunities the technology presents across the business groups.
bp’s AI strategy is targeting business value and has the potential to impact and increase efficiencies in every aspect of its business where there’s a profit center, from allocation of capital, to its enormous and complex supply chain which includes satellites and vessels for sea, highway, and rail transport, to predictive maintenance of assets, to peak optimization of wind turbines, to seismic interpretation of subsurface for hydrocarbon exploration and production that itself produces petabytes of data.
Also, importantly, analytics and AI will unlock value for marketing and in integrated supply and trading, bp’s energy trading arm, and the orchestration engine of the business that looks to enhance performance, manage risk, ensure compliance, and maximize value throughout the entire company.”
The right partner for measurable results
bp selected Cloudreach to help it achieve the goal of global optimization. “Cloudreach is widely recognized as a strategic partner for cloud engineering,” said Kimberly Anderson, Service Owner for Technology Incubation at bp. “With the type of scale and complexity we’re talking about, we needed a strong track record and some serious expertise in cloud engineering.”
bp’s AI strategy has three pillars: activate, accelerate, and advance. “Activate is around what our AI-enablement team is doing at the moment,” said Lance Goodship, Principal Architect for AI at bp, “building a foundation for the active adoption of AI across bp and a solid platform to get to the next stage. Accelerate is really all about rapidly scaling where we see innovation opportunities that add value. And in the advance phase, we will be taking products to a high degree of maturity by fully exploiting the capabilities of AI.”
Moving towards solutions at pace
For bp and its many other customers, Cloudreach is engineering Microsoft Azure analytics and AI solutions to help them leverage their data, become more efficient, and achieve strong outcomes for their business, including creating new lines of business.
Cloudreach deployed and configured a range of specialized Microsoft Azure services and features in bp’s upstream digital factory. “The spectrum of business challenges is enormous,” said Goodship. “We’ve found that practically every mature aspect of AI and machine learning, as well as every emerging one, has relevance in one of these applications across bp.”
With Cloudreach’s cloud engineering expertise, bp solved complex enablement challenges and was ready to implement Microsoft Azure AI and machine learning in just six weeks. “We’re building out automation patterns and accelerator templates,” said Anderson. “We’re implementing best practices and helping our data sciences community improve time to value.” bp, Cloudreach, and Microsoft are also working together to evaluate several Microsoft Azure services that are in private preview, enabling them in bp's environment to validate potential use cases.
Use cases the team has enabled include:
- Using Microsoft Azure Form Recognizer to extract data into a machine-readable format and developing an accelerator for bulk extraction for a region with a potential value realization of $20M+ USD.
- Creating a reusable pattern for automating the scaling and testing process allowing for unsupervised testing, which has the potential to reduce time to analyze a well from three months down to less than one week.
- Moving a solution from local desktop to Microsoft Azure, resulting in a reusable, scalable, and highly available solution with cost savings of over $2M USD.
- Currently evaluating Microsoft Azure Anomaly Detection and Microsoft Azure Project Bonsai (machine teaching), and how these services can provide value to the business.