United StatesEnergy and utilities2018 – ongoingRange Resources
Machine-learning well forecasts and development-scenario evaluation
Range Resources cut development-scenario analysis from days to hours
Problem
Engineers spent substantial time merging data and updating production forecasts. Evaluating changes to well designs also required estimates of production and financial returns.
Approach
Range partnered with Novi to develop software combining machine-learning forecasts with geoscience information and simulations. Engineers used it to compare completion designs and estimate their economics. Production forecasts used inputs updated weekly.
Results
Takeaway
Combining production forecasts with cost estimates let engineers assess development alternatives earlier. Range did not disclose a financial return on the software investment.
Categories
- Industry
- Energy and utilities
- Business function
- Production and qualityStrategy and organization
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Partner co-development
- Evidence
- Company's own claim
Sources
Source summaries
Range engineers described software developed with Novi for forecasting well production and comparing development options. They reported that assessing the economics of a major scenario change fell from days to hours, without disclosing an investment return.
Full details are in the original sources.