The growth in hydrocarbon extraction from shale basins has come with a set of challenges in doing so efficiently, safely and in an economically viable manner. The emergence of practices like Continuous Frac and Simul-Frac have become commonplace as operators continue to drive efficiency up, further increasing the frequency and complexity of decision-making throughout the job. Thus the impact of incorrect or non-timely decisions is accentuated, with unplanned downtime caused by unexpected equipment failure or human mistakes among the top reasons of Non-Productive Time. Hydraulic fracturing processes in North America have progressed majorly since its inception from a process once predominantly dependent upon human intervention towards more human-assisted automated operation.

In this paper, the authors have investigated the potential for providing AI-enabled assistance for the regular operational tasks as the solution for providing high centainty data-supported insights to support decision-making. Specifically, we looked at providing early alerts of probable equipment failures and flagging up process inefficiencies, or deviations, that may be subsequently addressed to further improve performance. We show the steps taken by authors and internal teams to successfully develop and deploy a digital control and assistant system within its hydraulic fracturing equipment offering that enables a shift from direct manual to semi- and fully autonomous operation, from reactive to a proactive approach to maintenance and failure prevention, all which in turn have resulted in prolonged operational life and efficiency of equipment, thus reducing Non-Productive Time.

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