Idea
ML-driven gas lift optimization platform boosting production efficiency in unconventional oil fields without costly downhole data.
Research Paper
Core Innovation
This paper introduces a novel ML model that forecasts the Gas Lift Performance Curve using only historical production time series data, eliminating the need for downhole gauges or multi-rate tests. Coupled with a Bayesian optimization framework, it efficiently determines optimal gas injection rates under facility constraints, enabling practical and scalable gas lift optimization in unconventional fields.
Why It Matters
Oil producers in unconventional fields face challenges optimizing gas lift due to lack of downhole data and costly testing. This workflow delivers measurable production uplift by automating gas injection optimization using only historical surface data, reducing operational costs and enabling scalable deployment across many wells. It transforms gas lift management into a data-driven, cost-effective process applicable to constrained facilities.
Market Size (TAM)
$2B–$10B TAM for oilfield production optimization; $500M–$1B SAM from unconventional oil producers. Driven by rising unconventional production and cost pressures to optimize lift methods.
Potential Customers & Pain Points
- Unconventional oil producers – Lack of downhole data limits gas lift optimization
- Oilfield service companies – High cost and complexity of multi-rate well tests
- Facility operators – Need to optimize gas injection within capacity constraints
Business Model
Subscription-based SaaS platform offering gas lift optimization as a service with tiered pricing based on number of wells managed and support levels.
Competitive Landscape
- Schlumberger
- Halliburton
- Baker Hughes
- Weatherford
Implementation Challenges
- Integration with existing field operations and control systems
- Data quality and availability variability across fields
- Operator trust and adoption of ML-driven recommendations
Validation Strategy
- Expand pilot deployments across diverse unconventional fields
- Demonstrate consistent production uplift and cost savings
- Collect operator feedback to refine user interface and integration
- Partner with oilfield service companies for broader market reach
Research Paper Overview
A Machine-Learning-Based Gas Lift Optimization Workflow for Unconventional Fields
Summary
This paper presents an automated ML-driven workflow to optimize gas lift injection rates in unconventional oil fields without requiring downhole gauges or multi-rate well tests. The workflow integrates a predictive ML model for gas lift performance and a Bayesian optimization framework to maximize production under facility constraints. Piloted on 30 wells in Bakken with over 5% production uplift, it is now deployed across 200+ wells, offering a cost-effective solution for fields lacking detailed downhole data.