Analyses have been widely applied in production forecasting of oil/gas production in both conventional and unconventional reservoirs. In order to forecast production, traditional regression and machine learning approaches have been applied to various reservoir analysis methods. Nevertheless, these methods are still suboptimal in detecting similar production trends in different wells due to data artifacts (noise, data scatter, outliers) that obscure the reservoir signal and leading to large forecast error, or fail due to lack of data access (inadequate SCADA systems, missing or abhorrent data, and much more). Furthermore, without proper and complete integration into a data system, discipline silos still exist reducing the efficiency of automation.
This paper describes a recent field trial conducted in Australia's Cooper Basin with the objective to develop a completely automated end-to-end system in which data are captured directly from the field/SCADA system, automatically imported/processed, and finally analyzed entirely in automated system using modern computing languages, modern devices incl. IoT, as well as advanced data science and machine learning methods. This was a multidisciplinary undertaking requiring expertise from petroleum, computing/programming, and data science disciplines.