Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Petroleum and Natural Gas Engineering

Committee Chair

Shahab D. Mohaghegh.


Generating long-term development plans and reservoir management of shale assets has continued apace. In this study, a novel method that integrates traditional reservoir engineering with pattern recognition capabilities of artificial intelligence and data mining is applied in order to accurately and efficiently model fluid flow in shale reservoirs. The methodology is efficient due to its relatively short development time and is accurate as a result of high quality history matches it achieves for individual wells in a multiwell asset. The technique that is named Artificial Intelligence (AI) Based Reservoir Modeling is a formalized and comprehensive, full-field empirical reservoir model. It integrates all aspects of shale reservoir development from well location and configuration to reservoir characteristics and to completion and hydraulic fracturing. This approach not only has a much faster turnaround time compared to the numerical simulation techniques, but also models the production from the field with good accuracy, incorporating all the available data. This integrated framework enables reservoir engineers to compare and contrast multiple scenarios and propose field development strategies. AI-based Modeling is applied to a Marcellus Shale asset that includes 135 horizontal wells from 43 pads with different landing targets. The full field AI-based Shale model is used for predicting the future well/reservoir performance, forecasting the behavior of new wells/pads and to assist in planning field development strategies. Furthermore, this study takes advantage of applying advanced pattern recognition tools in order to investigate the impact of design and native parameters on gas production as well as optimizing the completion and stimulation parameters for newly planned wells.