Abstract
In this study, the authors have analyzed well and production data beginning with more than 400 wells in the greater Sanish-Parshall area of the Bakken. The study used Geographical Information System pattern-recognition techniques along with other data-mining techniques to interpret trends in the data sets. The study was made possible by combining data sets from the North Dakota Industrial Commission Oil and Gas Division, public data, and in-house proprietary data.
The study was designed to search for relevant trends in the distribution of production results for wells completed with fracturing sleeves and packers, plug and perforated, or complex completions to determine whether differences in productivity existed and needed to be factored into completion recommendations.
Trends examined in the project in addition to completion type included treatment parameters such as fracturing fluid types and quantities, proppant types and quantities, number of completion stages and stage lengths, perforation cluster spacing and length, and calculated perforation friction drop. All parameters analyzed were examined for statistical importance.
This work is significant in that it shows that the application of practical data-mining methods to an intermediate-size Shale Oil (light, tight oil) well data set can result in learning key lessons that may not be apparent when working with small data sets. This work is significant in the use of merged reservoir quality proxies, well architecture data, completion data, and stimulation data, against which production results are placed in geographical perspective of the Bakken Formation for improved interpretation. The work is also significant in that it may be used to allow selection of completion systems on the basis of completion time and cost balanced against concerns over differences in well production impact of one system over another, e.g., frac sleeves versus plug and perf type and complex completion systems.
