COMMENTARY: AI Applications in Upstream Oil and Natural Gas – Part 2 – Yogi Schulz

By Yogi Schulz

Generative artificial intelligence (AI) is sweeping through the oil and natural gas industry. It’s out of control, like the Wild West. AI output is showing up in reports and presentations. The Apple App Store and Google Play offer many free AI apps of varying quality. Every AI software vendor provides access to their prompt website. AI output is part of search results. AI capabilities are integrated into desktop software.

By implementing AI applications, companies are transforming operations, from optimizing exploration and drilling to streamlining production and logistics. Through advances in machine learning, big data analytics, and automation, companies benefit from efficiency, safety, and environmental sustainability.

What are some practical applications where the oil and natural gas industry could speed up the adoption of AI? You will immediately recognize that none of these applications are new. The value that AI contributes to these applications includes:

More confident or accurate results.
Reduced elapsed time to achieve the results.
A significant reduction in staff effort to achieve the results.
Increased scope or scale in factors such as the number of entities such as wells or facilities, higher data resolution or larger geographic area.

This second article describes significant AI applications at the last two phases along the upstream exploration and production life cycle that optimize operations and reduce costs to increase margins. . The upstream life cycle consists of the significant phases:

Exploration
Field development
Well drilling and completion
Production operations
Abandonment

Consider which AI applications could bring the most value to your producer organization.

Production operations

Producers are engaged in an ongoing effort to reduce operating costs and unscheduled outages while increasing production and the percentage of volumes in place in the reservoir. AI improves data analytics and manages the additional data volumes to support production operations.

Production Optimization

It’s widely understood that about half of the production optimization investments that production engineers propose and are approved by management do not achieve their planned positive net return. Applying AI-driven analysis to production optimization can increase production volumes while containing or even reducing operating costs with a low failure rate. More specific production optimization examples include:

Optimizing rod lift crude oil pumps.
Simulating complex reservoir behaviour.
Predicting water encroachment or sand production to minimize downtime.

Production volume forecasting

Applying AI-driven forecasting to production volume time series can reduce the inconsistency or uncertainty associated with more traditional algorithms. The benefits include:

Improved cash flow forecasting accuracy.
Improved accuracy for pipeline nominations.
Ease of identifying wells that will benefit either from a workover or should be suspended.

Predictive maintenance

AI applications can quickly and accurately analyze large volumes of diagnostics data from Industrial Internet of Things () sensors at wells, gas processing plants, compressor stations, pipelines and terminals. The benefits include:

Improving pump and compressor performance.
Predicting failures and reducing unscheduled outages in production equipment, pipelines and ESPs.
Diagnosing root causes of equipment failures more quickly and accurately.

Supply chain management

Producers rely on a vast range of products and services to sustain their upstream operations. Some complex products require long lead times or specialty manufacturing processes. Those realities create reliance on a complex supply chain, especially when operations are located in poorly developed parts of the planet.

The benefits of AI-driven supply chain management include the following:

Reduced errors and out-of-stock situations by between 20 and 50%.
Significantly reduced risk of expensive development delays caused by avoidable out-of-stock situations.
Higher assurance that vendors are meeting industry-specific requirements for materials and tolerances.

Trucking optimization

Producers spend significant money on trucking produced water to disposal facilities, fresh water to well sites for hydraulic fracturing and oil and water emulsions to pipeline terminals. AI applications can improve route optimization because of the large number of variables involved, including:

Truck capacity and characteristics.
Terminal locations, limitations and hours of operation.
Road characteristics and load limitations.
Operator certifications for terminals.

The benefits of trucking optimization include the following:

Significantly reduced trucking costs.
Higher truck utilization.
Reduced waiting time charges at terminals.

Master data management

Master data management for structured data has been challenging to implement, despite significant benefits, due to the high effort for manual data analysis and correction. Automating master data discovery with AI to reconcile large amounts of geospatial and transactional data from multiple data sources offers a considerable improvement. The result is high-quality data in a single datastore that’s easy to query. The benefits of AI-supported master data management include:

Making reliable structured data available for decision-making.
Significantly reduced manual data review efforts by data stewards.
Increased confidence in query results leading to higher confidence in recommendations.
Laying the digital foundations for end-user tools and applications.
Reduced query effort for company staff.

Unstructured data management

Unstructured data management has been challenging to implement despite significant benefits due to the following issues:

Difficulty identifying available data sources.
High effort to select and manage metadata.
Cost of scanning, digitizing and tagging large volumes of paper records.
Difficulty of harmonizing various terminologies.

Addressing these issues with AI offers opportunities to unlock the value of unstructured data. The result is high-quality digital data that’s easy to query. The benefits of AI-supported unstructured data management include:

Significantly reduced manual data review efforts by data stewards.
Ability to easily access this formerly inaccessible corporate data for decision-making.

Abandonment

All producers are concerned about reducing the cost and distraction of well, facility and pipeline abandonments. AI applications can improve abandonment designs to address the following situations:

Risk of insufficient planning due to missing historical data about downhole equipment, drilling events, production problems, well logs and geological data.
Potential to share mobilization costs through cooperation with neighbouring producers.
Risk of discovering problematic downhole situations such as casing and cement deterioration after starting the work.
Risk of groundwater contamination.

The benefits of AI-driven abandonment designs include the following:

Reducing the risk of cost overruns.
Reducing the risk of safety incidents.
Increasing the quality of environmental remediation.
Ensuring the adequacy of documentation and reporting for regulatory compliance.

AI applications offer oil and natural gas producers an enhanced ability to increase revenue, reduce cost and control risk in all phases along the upstream exploration and production life cycle.

Yogi Schulz has over 40 years of experience in information technology in various industries. He writes for , , and other trade publications. Yogi works extensively in the petroleum industry to select and implement financial, production revenue accounting, land & contracts, and geotechnical systems. He manages projects that arise from changes in business requirements, the need to leverage technology opportunities, and mergers. His specialties include IT strategy, web strategy, and systems project management.