Sustained 10x ROI for a Major Operator
Modern oil and gas operations generate enormous amounts of data, but inefficiencies in analyzing and acting on this data can lead to significant production losses. Tasq is an AI powered operational platform designed to bridge this gap by automating production workflows and improving decision making. Deployed at a major operator, Tasq integrated into control rooms, assisted production engineers, and coordinated field operations to create a data driven ecosystem. The result was a transformation from a manual, reactive workflow to a proactive, AI driven approach, improving efficiency, production output, and cost savings. Tasq was implemented across all wells in the basin with multiple artificial lift types and well types. Tasq connected 5 different data sources across SCADA, Artificial Lift, Production Accounting, Downtime, and WellView all into 1 system.
Key Results
- Immediate $1M increase in production revenue within the first month
- 10x ROI in the first month, scaling to a sustained 100x ROI, with the organization internally reporting a 2,000 bbl/day uplift ($36M/year)
- 2x improvement in downtime reporting, uncovering $75M/month in previously untracked opportunities
- 30% reduction in downtime, aligning with the primary objective of automating downtime reporting and enhancing operational efficiency
- Streamlined workflows leading to faster issue resolution and a 3x increase in operational efficiency
Problem Statement
Key challenges: Wasted time daily from poor data quality and manual data entry. Unreported downtime that missed small yet cumulative losses. Fragmented software systems requiring excessive manual consolidation. Reactive decision making. High operational costs from workflow inefficiencies.
The next step in operational efficiencies is changing from logic based SQL to learning systems. This is the single most important step in creating the next level of operational efficiencies. It is key that an organization know each production issue at a very detailed level, to route the right work to the right people.
Tasq AI Models
Tasq deployed five AI models to automate the flow of work within the control room, engineers, and field operations.
Well Target Model: ML model that dynamically updates each wells targets without human intervention. Result: 2x more production opportunity uncovered.
Bulk and Test Prediction Model: ML model trained to project gas and oil rates from the wells pressure data, when the well is not in test. Result: 30% more deferment uncovered than current reporting.
Setpoint Optimization Model: Setpoint changes are recommended to optimize a well, providing value in production, while reducing human diagnosis time. Result: 11% production increase for wells that accept recommendations.
Real Time Model: Tasq catches underperformance live as it is happening. Each of these are actionable and yield a live prioritization list.
Build Your Own Model
Build models 200x faster with Tasq. Examples of models created by the organization: Separator Burner Issue Model, Gas Lift Paraffin in Tubing Model, Dump Failure Model, Tank Leak Model, Arrival Sensor Model.
Downtime Reduction
Tasq's downtime automation provides an accurate, real time picture of downtime events, eliminating the inefficiencies of manual reporting. Comparing Tasq's automated downtime tracking vs. manual client reported downtime, Tasq uncovered 2x more production than what was previously being recorded. Tasq AI models identify 4x more production opportunities above what was being reported. Tasq reduced downtime by 30% vs previous baseline.
Conclusion
Key Takeaways:
- Automated Downtime Detection: Tasq uncovered 2x more downtime events and led to a 30% reduction in downtime over six months
- Operational Efficiency: Operators no longer waste time manually entering downtime data
- Significant ROI: $1M in new production revenue in the first month. Sustained 10x ROI scaling to 100x ROI. 2,000 bbl/day production increase equating to $36M/year
- AI Driven Decision Making: Teams work from a single source of truth, eliminating guesswork