Developing a Predictive Model to Minimize Gel Defects in Polyethylene Film Production

This Praxis introduces the Supervised Machine Learning Algorithm Performing Quantitative Statistical Analysis of the historical data from film surface analyzer (FSA) as a new methodology to identify the root causes of variation of defect rates for each process run. This study aimed to identify significant contributors to gel formation, develop a predictive model, and show that using a predictive machine learning model using FSA data enhances the prediction of gel defects. The primary goal of using the predictive model is to improve product quality. The model’s input considers attributes such as product grades, catalyst types, and process variables/tags (Min, Avg, Std, Max). To minimize the cost and maximize the profit of a petrochemical plant, gel defects should be mitigated or eliminated via an optimum technical solution. The predictive model will help research engineers and operations detect the film defects in advance and prevent the gel defects as much as possible. The predictive model may also help to clarify the characteristics of gels seen in film-based products.

Author

• Ahmadian, Hassan

Language

• en

Date created

• 2023

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Blackburn, Timothy

Committee Member(s)

• Islam, Muhammad
• Etemadi, Amir

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/p2676w29s