Various datasets are generated during hydraulic fracturing, flowback- and well-testing operations, which require consistent integration to lead to high-quality well performance interpretations. An automated digital workflow has been created to integrate and analyze the data in a consistent manner using the open-source programming language R. This paper describes the workflow, and it explains how it automatically generates well performance models and how it analyzes raw diagnostic fracture injection test (DFIT) data using numerical algorithms and Machine Learning. This workflow is successfully applied in a concession area located in the center of the Sultanate of Oman, where to date a total of 25+ tight gas wells are drilled, hydraulically fractured and well-tested. It resulted in an automated and standardized way of working, which enabled identifying trends leading to improved hydraulic fracturing and well-testing practices.

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