The Most Common Errors Used to Create Type Wells
November 22, 2016
Randy Freeborn, P. Eng., Chief Research Engineer for Energy Navigator
Type Curves, Decline Analysis, Oil and Gas Forecasting, Reservoir Engineering
Studies have shown that the most common method for creating type wells for use in forecasting wells consistently leads to the overestimation of reserves. Some studies have shown these results to be off by as much as 25%. My work with hind casting studies shows those type wells are more likely to overstate EUR, and by an amount as high as 40%.
This becomes a problem because these errors are used to guide important capital decisions within an organization, and can lead to poor allocation of resources or decreased investor confidence. When times are good, companies can overcome such bad projections with higher pricing. But in times like these, overspending on capital projects like plants or extra drills could result in bankruptcy. It’s time to look at these common errors inherent in the standard way of doing type wells and learn how to avoid them.
Forecasting groups instead of grouping forecasts
When only using history, we average representative wells first and then forecast from the average – I call this forecasting groups and it can affect forecast integrity.
Instead, we should first forecast all the wells individually, and then average – in other words, we should group forecasts, not forecast groups. By forecasting each well first, we are given more opportunity to see trends develop. Forecasting from an averaged group tends to mask trends. Forecasting individual wells has the added advantage of having forecast differences in some wells offset the differences in other wells.
In the example above, I hind casted 9 wells that produced for 27 continuous years. I truncated the history to 8 years to build a type well, and used the remaining 19 years to confirm the results. The white line shows the type well one would choose using the common method of forecasting from the grouped results. The blue shows the actual data for the full 27 years. The yellow line shows the type well one would choose using the recommended grouped forecast method. As you can see, there is a tremendous difference between the effectiveness of the two methods.
Depleted wells and declining well counts
The other error in creating type wells results from shifting the time scale before averaging, so that all wells start producing at the same time. At various times, wells will no longer have data to average; either because they have been depleted or have produced for fewer months than the other wells. In both cases, there is a “Survivor Bias” or focus on the wells that survive.
There is an unintended consequence when the well count is reduced, which is that the type well rate behaves as though the depleted or shorter history well continues to produce at the average rate. When that happens, you get a spike or drop in the type well due to the shrinking number of wells in the sample.
In the example above, the lowest producing well has a shorter production history than the other two. So when it drops out, the common method for type wells would result in a type curve rate that jumps up, as the rate would be divided by the better top two producing wells. This is the more common outcome and is the result for depleted wells or a best well first strategy. If performance improvements span a long time period, better wells may run out of production resulting in a decrease in the type well rate.
To resolve the error, every well must have a rate – so a depleted well continues to produce at a rate of zero. For producing wells, using your forecast rate instead of production history for the basis of a type well will eliminate the issues of comparing wells with 7 years of history to wells with 4 years of history. In the example above, the third well’s decline would be taken into account for the entire type well, leading to a more accurate curve.
Here’s a simple example to illustrate survivor bias. Three wells are used to create a type well, which will in turn be used to predict the performance from three undrilled wells. Both should have the same total field rate. The type well rate is the field rate divided by the number of wells with data to average: 3 for 2 months, then 2.
To use the type well, we multiply the type well rates by the 3, the number of wells we plan to drill. We do not multiply by 3 until the end of month 24, then multiply by 2 because the number of wells in the type well decreased. Comparing yellow numbers in the “drill 3 wells” row to the “field total” row, demonstrates that the common method of creating a type well inherently assumes that a well without production will produce at the type well rate. Clearly, that’s not logical and can result in an overstatement of EUR.
The Answer: Forecast Wells First
These are errors inherent in the common method of doing type wells. Not mistakes made by bad engineering, but rather flaws in the process itself. The solution for all of them is to change how you do type wells: forecast EVERY well first, then group those forecasts together to make your type well. Doing so will better identify trends in ALL wells in the sample and remove the errors associated with survivor bias.
In the past, it wasn’t feasible to forecast every well, which is why the common method became the standard method. Concessions were made due to the labor-intensive methods for creating forecasts of the past. Modern software solutions make forecasting wells much faster, thereby allowing engineers to create better, more reliable type wells when they forecast first, then average the history and the forecast.
About the author:
Randy Freeborn is a Distinguished Lecturer of the SPE on the subject of type wells and a subject matter expert in empirical forecasting and related technology. Currently, he is Chief Research Engineer at Energy Navigator where he is responsible for identifying and inventing engineering technology for inclusion in our Value Navigator software. He has been a professional engineer for 43 years and is a member of SPEE and SPE. He has given guest lectures at the University of Houston and Texas A&M, and has been called as an expert witness.