A Forecast Model for Classification for Cost Components in Midstream Oil and Gas Pipeline Construction

In the context of a growing global pipeline market, the escalating demand for oil and gas infrastructure is met with intensified scrutiny over environmental impacts, necessitating much-needed attention to key cost components during project planning and evaluation to not only pass the Federal Energy Regulatory Commission (FERC) certification necessary to start construction but also to consider the initial uncertainty that can be threatening to the financial security of construction companies servicing the midstream segment of the oil and gas industry. This praxis research presents a quantitative methodology to classify the total construction costs of such projects, employing a stacking ensemble technique that combines prediction outputs of cost components as the base models and then uses those predictions to classify the total construction cost. Through testing and analysis of various classification algorithms, the research achieves an average accuracy rate of approximately 82%, highlighting the potential of advancing the body of knowledge of cost estimation using a stack ensemble classification model. The findings underline the critical need for high-quality, reliable data, suggesting that future research could greatly benefit from an expanded dataset to refine further and improve cost estimation methodologies. This need continues to be a priority with every new pipeline construction under development. Lastly, this approach addresses the immediate challenges faced by the industry and lays the groundwork for more sustainable and economically viable pipeline construction projects in the face of an evolving environmental landscape.

Author

• Solichin, Muhammad Afrinal

Language

• English

Keyword

• Cost Overrun
• Machine Learning
• Stacked Generalization
• Out-of-Fold Predictions
• Stack Ensemble Classification Model
• Quantile Classification

Date created

• 2024

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Ali, Daniyal A.

Committee Member(s)

• Ali, Daniyal A
• Fong, Titus
• Blackford, Joseph P

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/r494vm219