Reliability Solutions
Tested and Proven

Learn how the Open AI Energy Initiative can improve the performance and productivity of energy assets and processes, supported by leaders in energy technology

Learn More

Case Studies

OAI Reliability Solutions have been implemented at scale by the world’s largest energy companies. Learn how AI-driven and domain-specific OAI Reliability Solutions can improve the performance and productivity of energy assets and processes.

Reduce Unplanned Downtime

Problem

One of the largest global oil and gas companies in the world lacked a reliable solution to predict failures of electrical submersible pumps (ESPs) for high-value wells. Unplanned ESP failures resulted in lengthy unplanned ESP replacements and several weeks of lost production per incident.

​Solution

To reduce unplanned well downtime and implement a more proactive ESP replacement strategy, the company selected BHC3 Reliability. Within 14 weeks of project kickoff, the Baker Hughes and C3 AI teams were able to ingest 10 years of historical data from 16 disparate source systems, build 1,500 supervised and unsupervised ML models, and deploy a production-ready AI machine learning application for 100 offshore wells across several fields in two countries. BHC3 Reliability identified high-risk wells up to 60 days before an ESP failure with recall of 66%, significantly reducing maintenance costs and production downtime.

Benefits

With BHC3 Reliability, the company can achieve over €90 million of additional annual revenue from increased production and reduced costs.

Reduce Unplanned
Downtime

Predict Unplanned
Outages

Predict Unplanned Outages

Problem

One of the largest North American gas pipeline companies lacked a reliable solution to predict unplanned gas generator outages in compressor stations. When a compressor unit shuts down, natural gas throughput of the pipeline network is reduced, impacting the company’s profitability and quality of service to its customers.

​Solution

The company selected BHC3 Reliability to reduce unplanned compressor unit outages. Within 12 weeks of project kickoff, the Baker Hughes and C3 AI teams were able to ingest 5 years of historical data from six source systems, build over 4,000 ML model configurations and deploy a production-ready AI machine learning application. BHC3 Reliability reduced the number of false alerts by 99% and predicted 52% of unplanned gas generator outages up to 48 hours in advance.

Benefits

Using BHC3 Reliability, operators can identify high-risk compressors in advance and dispatch resources before a shutdown. With BHC3 Reliability, the company can achieve over $80 million of additional annual revenue from reduced downtime and maintenance costs.

Increase Annual Savings

Problem

A major hydrocarbon producer committed to increase its utilization of associated petroleum gas, which is transported by mobile gas compressors. Within a few months of installing these compressors at one of its fields, operators were overwhelmed with 1,500 alarms per month from each compressor and daily compressor failures, leading to unplanned shutdowns and increased maintenance costs.

​Solution

To improve the quality of alarms and accurately predict compressor failures, the company selected BHC3 Reliability. Within 10 weeks, the Baker Hughes and C3 AI teams ingested over 55 million rows of data from four disparate data sources, tested over 70 features, and built over 50 ML model configurations. BHC3 Reliability accurately predicted 50% of unplanned shutdowns with over 22 hours of lead time, reduced false alarms by 99% and reduced non-productive time by 16%.

Benefits

With BHC3 Reliability, the company can achieve more than $40 million in annual savings from increased productivity, increased gas throughput, and reduced maintenance costs.

Increase Annual
Savings

Identifying the Cause
of Erratic Equipment
Performance

Identifying the Cause of Erratic Equipment Performance

Problem

The operator of a deepwater production platform wanted proactive ways to discover and mitigate control valve failures before they occurred.

​Solution

The team selected the Shell Predictive Maintenance for Control Valves application on the BHC3 AI Suite to deploy a scalable predictive maintenance solution. Using this solution, which has proven its value across multiple Shell assets, the oil & gas company can implement machine learning models at scale to predict the expected behaviour of the control valves at the upstream asset. The application flags any anomalous behaviour to remote or onsite engineers who can pre-emptively address failures.

Benefits

With Shell Predictive Maintenance for Control Valves, the operator diagnosed hunting behaviour in a process-critical temperature control valve within a few weeks of installing the solution, extended the working life of the control valve, enhanced process stability, avoided pressure and/or temperature related trips to the main gas compressor and increased production by reducing unplanned downtime due to valve failure.

Download case study

Fault Finding for Rotating Equipment on an Upstream Platform

Problem

The operators at one of Shell’s largest refineries wanted a proactive way to discover and mitigate control valve failures before they occurred.

​Solution

The team selected the Shell Predictive Maintenance for Control Valves application on the BHC3 AI Suite to deploy a scalable predictive maintenance solution. Using this solution, the oil & gas company can predict the expected behaviour of control valves at the downstream manufacturing asset. The software flags anomalous behaviour to remote or onsite engineers who can pre-emptively address failures.

Benefits

With Shell Predictive Maintenance for Control Valves, the oil & gas company prevented margin loss due to shutdown of the downstream water stripper unit and production loss, avoided the risk of flaring and associated emissions and protected customer reputation by ensuring that erratic performance in the main fractionator did not result in poor quality product.

Download case study

Fault Finding for Rotating
Equipment on an Upstream
Platform

Diagnosis for a
Process-Critical
Temperature Control Valve

Diagnosis for a Process-Critical Temperature Control Valve

Problem

The operator of a high-pressure, high-temperature oil and gas field wanted proactive ways to discover and mitigate control valve failures before they occurred.

​Solution

The team selected the Shell Predictive Maintenance for Rotating Equipment application on the BHC3 AI Suite to deploy a scalable predictive maintenance solution. This solution would enable the team to predict the expected behaviour of centrifugal compressors at this upstream asset. The software flags anomalous behaviour to remote or onsite engineers who can pre-emptively address failures.

Benefits

Using Shell Predictive Maintenance for Rotating Equipment, the oil & gas company avoided the need for unplanned shutdown of the platform’s gas compressors, potentially prevented the risk of unnecessary flaring for maintenance from this event, identified the problem more quickly than a conventional fault-tracking system and and enabled effective maintenance planning for major equipment.

Download case study

“We are excited to take Shell's proven predictive maintenance solutions to market and want to develop an open ecosystem where others can offer AI solutions to help improve reliability across the industry.”

Yuri Sebregts Chief Technology
Officer

“Working alongside our alliance partners at C3 AI and together with industry leaders at Shell and Microsoft, the OAI will help address the persistent industry challenge of nonproductive downtime.”

Uwem Ukpong Executive Vice President of Regions,
Alliances and Enterprise Sales

“The AI and data aggregation capabilities of the BHC3 AI Suite and BHC3 Reliability, combined with OAI’s domain-specific reliability solutions will increase operational availability and efficiency.”

Ed Abbo President and
Chief Technology Officer

Request Briefing

By submitting your information, you agree to our Privacy Policy and Terms and Conditions.