Correlating Geological and Seismic Data with Unconventional Resource Production Curves using Machine Learning


We present a machine learning workflow that incorporates geophysical (seismic attributes) and geological data, as well as engineering completion parameters, into a model for predicting well production curves. The study area is in southwest Texas in the lower Eagle Ford formation. We make use of functional principal component analysis to summarize the well production time series, and random forest models to predict the functional principal components with geological, geophysical and engineering data as predictors. We are then able to predict the well production time series, with 65-76% accuracy. This method incorporates disparate data types into a robust, predictive model that predicts well production in unconventional resources.

Meeting Name

88th Society of Exploration Geophysicists International Exposition and Annual Meeting, SEG 2018 (Oct. 14-19, Anaheim, CA)


Geosciences and Geological and Petroleum Engineering

Research Center/Lab(s)

Center for Research in Energy and Environment (CREE)

Second Research Center/Lab

Center for High Performance Computing Research

Keywords and Phrases

Artificial intelligence; Decision trees; Forecasting; Geology; Geophysical prospecting; Learning systems; Resource valuation; Seismology; Time series; Time series analysis, Engineering data; Functional principal component analysis; Predictive modeling; Principal Components; Production curve; Seismic attributes; Unconventional resources; Well production, Principal component analysis

Document Type

Article - Conference proceedings

Document Version


File Type





© 2019 Society of Exploration Geophysicists (SEG), All rights reserved.

Publication Date

01 Oct 2019