Failure Prediction of Pvc Transmission Pipes Using Data-Driven Modeling | IConEST

Paper Detail

Title

Failure Prediction of Pvc Transmission Pipes Using Data-Driven Modeling

Authors

Dr. Thikra Dawood, Purdue University, United States of America
Assoc. Prof. Dr. Emad Elwakil, Purdue University, United States of America
Lecturer Hector Mayol Novoa, University of St Augustin of Arequipa, Peru
Lecturer José Fernando Gárate Delgado, University of St Augustin of Arequipa, Peru

Abstract

Polyvinyl chloride (PVC) pipes are currently used in 80% of the water infrastructure in North America. Failure incidences in PVC pipes occur because of poor manufacturing practices, improper installation, or third-party damage. Data-driven modeling techniques have been extensively used in solving water infrastructure problems, especially when the obtained data are limited. this paper aims at developing a data-driven model that applies computational algorithms and provides the analytical foundations to predict future PVC pipe failures. First, data related to hydraulic pressure and flow rate are collected from the City of El Pedregal in Peru. The data are streamlined and fed to the simulation and regression machine. After successive simulation trials and numerous polynomial functions, the best fit model is selected. The coherency of the model is investigated through different statistical metrics, in conjunction with residual analysis. The results indicated the efficacy of the model with mean absolute error (MAE) of 0.35. The developed model is a predictive tool that can be used by infrastructure managers as a preemptive measure against future pipe breaks or leakages.

 Keywords

failure prediction, data-driven modeling, pvc pipe, water infrastructure, regression analysis.  

Citation

Dawood, T., Elwakil, E., Novoa, H.M. & Delgado, J.F.G. (2020). Failure Prediction of Pvc Transmission Pipes Using Data-Driven Modeling . In M. Shelley & I. Sahin (Eds.), Proceedings of IConEST 2020--International Conference on Engineering, Science and Technology (pp. 42-47). Monument, CO, USA: ISTES Organization. Retrieved 15 January 2021 from www.2020.iconest.net/proceedings/10/.

Links

Download Fulltext
Announcements

Web of Science - Conference Proceedings Citation Index (CPCI)

The proceedings of our collaborative conference, Proceedings of International Conference on Social and Education Sciences (IConSES)-2019, is selected by Web of Science for coverage in the Conference Proceedings Citation Index (CPCI). The Proce...

May 15, 2020

View details »

Annual Book Publication

The Studies on Engineering, Science and Technology 2020 (SonEST2020) is a peer-reviewed scholarly online book. The invited papers are reviewed by at least two international reviewers with expertise in the relevant subject area. The book is a refereed book and has a double-blind review. ...

March 14, 2020

View details »

Information about Virtual Presentations

Dear Participant, We will use the Adobe Connect Meeting platform for the virtual presentations. You will receive an email titled “Adobe Connect - Your Account Information” from Adobe for the Account URL link and password to participate in the vi...

November 06, 2019

View details »

Supported by
IOWA
Indiana University
University of Northern Colorado
International Society for Technology, Education and Science
Participating Countries ICONEST 2020

Abstracting/Indexing

The publications affiliated with ISTES Organization are indexed or listed by all or some of the following sources:

Sponsors