Global Oil & Gas Leader Leverages the BHC3 AI Suite
One of the world’s leading oil and gas producers looked for a better way to reduce costly equipment failures at its assets around the globe. Each of this company’s businesses is composed of multiple assets that include oil well basins, offshore platforms, refineries, pipelines, and retail outlets. Each asset operates vast arrays of equipment that are orchestrated with the goal of maximizing the efficiency of oil production. At any point in time, any one of these components can fail – with potentially catastrophic results.
With millions of assets at different locations around the world, creating a scalable predictive asset maintenance solution posed a significant challenge from both an architecture and data science perspective. After exploring alternative approaches, the oil and gas producer selected the BHC3 AI Suite, the leading enterprise AI platform for oil and gas brought to market jointly by Baker Hughes and C3 AI. It was chosen as their AI platform of choice to build scalable, enterprise software enabled by AI–beginning with predictive maintenance use cases. Working collaboratively with BakerHughesC3.ai experts in a purpose-driven Center of Excellence (CoE), the oil and gas producer has moved swiftly by leveraging the capabilities of the BHC3 AI Suite to create and deploy BHC3 Reliability applications at enterprise scale within months of kick-off. These applications notify instrument engineers when asset components are behaving abnormally. This supports a proactive approach to maintenance in which engineers prevent failures before they happen.
About the Global Oil and Gas Company
- Hundreds of billions of dollars in annual revenue
- Millions of barrels of oil equivalent per day
- Tens of thousands of employees
Operations are divided into different businesses within theupstream, midstream, and downstream value chain
Rapid Software Development at the Center of Excellence
BakerHughesC3.ai worked with the oil and gas producer to establish a global Center of Excellence with two development teams spread across four different countries. The CoE provides the oil and gas producer with a governance structure and dedicated BakerHughesC3.ai personnel to share best practices, identify obstacles, and keep work on schedule. The BakerHughesC3.ai and customer teams collaborate to identify high-value use cases, specify and build applications within the BHC3 AI Suite, and run and optimize them over time. In the CoE’s first year, the oil and gas producer developed two applications now in production and one deployed for field testing. They now aim to deliver more than 15 distinct applications over a period of four years.
Predicting Failures for Control Valves Across the Globe
BHC3 Reliability for Control Valves enables instrument engineers to perform predictive maintenance at scale to all control valves within the asset. BHC3 Reliability is composed of a blend of predictive maintenance and machine learning model management. It is accessible to end users via a web UI application interface built entirely using the BHC3 AI Suite’s model-driven architecture.
Predicting Failures for Compressors Across the Globe
BHC3 Reliability for Compressors enables instrument engineers to perform predictive maintenance at scale on all compressors within the asset. The application leverages a non-linear machine learning model custom-tailored for each compressor by the oil and gas producer’s data scientists and subject matter experts. It is accessible to end users via a web UI application interface built using the React framework and interfaces with the BHC3 AI Suite via REST API.
Predicting Failures for Progressive Cavity Pumps
Field trials of BHC3 Reliability for Progressive Cavity Pumps (PCPs) are now in progress at one of the oil and gas producer’s natural gas divisions. These pumps were a leading cause of well failures within the division and the second largest annual operational spend. Overall, the estimated annual economic impact of PCP failures across all 3,000 of the division’s wells is more than $60 million.