The quality and quantity of information available in the public domain is growing rapidly, and companies are generating inhouse databases to track this explosion of information to improve operating and service performances. The hardware and software used to obtain and manipulate massive amounts of information are constantly improving. All these events have created an opportunity to evaluate the complex interaction of variables and quantify how they relate to a required result.
The subject of this paper is an analysis of granite wash completions in the Red Deer Creek field, Roberts County, TX. The analysis uses an artificial neural network (ANN). Specific areas of interest include any controllable/quantifiable aspect of a well's completion and stimulation procedure—including fluid selection, treatment volume, proppant type and volume, pump rates, and perforation distribution—that affects production outcome. Relevant conclusions are drawn that quantify the effects of reservoir, well, and completion factors on production results. This paper will document a test of methodology on the completion of a new well that resulted in a two-fold gas production increase compared to four previous completions that used only conventional completion optimization techniques.