Degree

Master of Science in Data Science

Department

Department of Computer Science

Faculty/ School

School of Mathematics and Computer Science (SMCS)

Date of Submission

Spring 2026

Supervisor

Dr. Muhammad Atif Tahir, Professor and Program Coordinator, Graduate & Postgraduate Programs (CS)

Committee Member 1

Dr Atif Tahir

Keywords

Shale Gas, Machine Learning, Cat Boost, GBRT, Well Design, Production Fore casting, Marcellus Shale

Abstract

This study presents the integrated, ensemble based framework that simultaneously predict the Estimated Ultimate Recovery (EUR) of shale gas wells and also tells how to design the wells in order to enhance the overall prediction. In this study, we focus on the Marcellus shale reservoir and analyze parameters like density, porosity, normalized gamma ray, resistivity, perforation interval length, proponent thickness etc. These features vary across different region of the shale reservoir. Multiple regression based approaches were evaluated while satisfying the need for the bias variance trade off. Among all the approaches the GBRT performed the best with the R2 of 0.85. The results depicts that there is the non-linear relationship among the features of the dataset. Overall, the study contributes to the efficient reservoir management that guarantees the maximum production white doing the resource management efficiently.

Document Type

Restricted Access

Submission Type

Research Project

Download

The full text of this document is only accessible to authorized users.

Share

COinS