Type Curves Part 7: Survivor Bias
Editor’s Note: While VISAGE rebranded to VERDAZO in April 2016, we haven’t changed the VISAGE name in our previous blog posts. We’re proud of our decade of work as VISAGE and that lives on within these blogs. Enjoy.
A common issue with Type Curves is that they are “Survivor Biased”, and can provide an unrealistic (optimistic) production outlook.
Survivor Bias Definition: as depleted wells are excluded from the monthly average-production-per-well calculation, the Type Curve values are biased by the surviving wells (i.e. wells that are still producing). As such the Type Curve does not reflect the actual average production one might expect taking into account wells that are depleted (or shut-in permanently).
This chart illustrates how Survivor Bias can cause latter-life production increases in a Type Curve.
Few software products provide Survivor-Bias Controls that allow you to include zeros in the production average after wells are identifiably depleted. The challenge when dealing with public production data is how to identify that a well is depleted. At VISAGE we chose to use a “period of non-production” as the mechanism to identify a depleted well. The user can define the length of this period (e.g. if a well has not produced in the last 12 months, then consider it to be depleted … meaning it is not expected to come on production again).
The following chart illustrates how Survivor Bias Controls can provide a more realistic expectation of production when depleted wells are included in your Analogue Well Selection.
Challenges of the Current Commodity Price Market
The current commodity price market will make identifying depleted wells more complicated. Increasingly companies are shutting-in wells until commodity prices recover to an economically acceptable level. This means that a larger window of non-production may need to be specified to identify a “depleted” well. Regardless, this is a poignant reminder that careful attention should be made to your Analogue Selection and every aspect of the choices you make when generating your Type Curve.
That concludes part 7 of this series. The remaining topics that you can look forward to include:
- Truncation Using Sample Size Cut-off
- Forecast the Average vs Average the Forecasts
- Representing Uncertainty
Production data: IHS Information Hub
Thanks for reading. We welcome your questions and suggestions for future blogs.
Some other blogs you may find of interest:
- Type Curves Part 1: Definitions and Chart Types
- Type Curves Part 2: Analogue Selection
- Type Curves Part 3: Normalization
- Type Curves Part 4: Calendar Day vs Producing Day
- Type Curves Part 5: Condensing Time (Idealized Type Curves)
- Type Curves Part 6: Operational/Downtime Factors on Idealized Curves
About VISAGE – visual analytics for the petroleum industry VISAGE analytics software equips operators and analysts in the petroleum industry to make the most valuable and timely decisions possible. VISAGE brings together public and proprietary oil and gas data from multiple sources for easy to use interactive analysis.