Unconventional plays have considerable uncertainty in production and economics. This paper proposes Decision Entropy Theory to represent uncertainty in unconventional plays in an objective and defensible way. The motivation to develop this theory is to account for the possibility of events occurring that are beyond our range of experience. This theory characterizes uncertainty in the context of making a decision; the case of maximum uncertainty corresponds to the maximum entropy for the possible outcomes of the decision. Assessing the value of test wells before committing to developing a play with unconventional resources is used to illustrate the theory.

We pose the decision to develop an unconventional play in terms of the cost of development, the ratio of the return-on-investment for productive wells in the play, and the uncertain frequency of productive versus unproductive wells in the play. Results for the value of tests wells are obtained and provided graphically in non-dimensional relationships.

Significant conclusions are the following:

  1. The probability distribution for the uncertain frequency of productive wells is important both for deciding whether or not to develop a play and for the value of information from test wells.

  2. It is unreasonable to assume that the probability distribution for the frequency of productive wells can be established based entirely on experience because that precludes the possibility of events beyond our experience. Such events can be particularly important with unconventional plays where experience tends to be limited.

  3. The value of test wells is enhanced when leaving open the possibilities for profit on a play with a small return-on-investment ratio and the possibility for loss on a play with a large return-on-investment ratio.

  4. There is a balance between relying entirely on historical data from analogous plays versus not relying on them at all. Within this balance, information for test wells at a new play can be combined with information from analogous plays to inform the decision about developing the new play.

  5. The value of test wells in a new play depends on the break-even frequency of good wells in the play; in the case of no information, this value is a maximum when the break-even frequency is 50 percent.

You can access this article if you purchase or spend a download.