A major challenge faced by the oil and gas industry is estimation of reserves for unconventional reservoirs. Traditional methods used to predict reserves for conventional wells are inappropriate for unconventional wells.
Construction of "Type Well" profiles is widely considered to be a useful tool for production forecasting, especially in presentations to investors. Type wells play an important role in estimating the EUR for a particular new well and for undrilled wells. They are responsible for guiding forecasts in wells expected to have similar behavior and characteristics. The conventional approach is to determine the arithmetic average of production during a given month from different wells in a reservoir of interest to create a type well. This methodology is deeply flawed, however, and results in significantly different estimates by different evaluators. The conventional approach may lead to either overestimates or underestimates (usually overestimates) of future production from a particular well or reservoir. This paper discusses the problem in depth and recommends possible ways in which type well construction can be made more realistic. In order to ensure more accurate forecasts from an individual well, a new approach is proposed. A detailed procedure to identify and remove outliers from production data is implemented in the workflow. We analyzed production data from Barnett shale reservoir and incorporated our findings in our proposed method to create type wells for a wide variety of reservoir types, with variations in both rock and fluid characteristics. Our analysis also includes suggestions for applying type well production profiles generated in one portion of a given reservoir to another area in the reservoir with different geological characteristics, notably permeability.