Performance comparison for pipe failure prediction using artificial neural networks

S. Kerwin, Borja Garcia de Soto, B. T. Adey

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Infrastructure managers must decide on the replacement timing of buried pipes in water distribution networks. These assets deteriorate resulting in failures and their associated consequences. In recent years, studies have investigated failure prediction models for pipes based on artificial neural networks (ANNs). These models are either generalized (i.e. trained with all pipe failures) or specialized (i.e. trained with certain pipe failures based on pipe material or failure history). It is currently unclear whether prediction accuracy is improved by developing several specialized ANNs compared to a single generalized one. To answer this question, four modelswere developed: a generalized model for cast iron (CI) and ductile iron (DI) pipes; a specialized model for CI pipes, a specialized model for CI pipes with no previous failures and a specialized model for CI pipes, which had previously failed. Overall, the study found minimal difference in performance between the generalized and specialized models.

Original languageEnglish (US)
Title of host publicationLife-Cycle Analysis and Assessment in Civil Engineering
Subtitle of host publicationTowards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018
EditorsDan M. Frangopol, Robby Caspeele, Luc Taerwe
PublisherCRC Press/Balkema
Pages1337-1342
Number of pages6
ISBN (Print)9781138626331
StatePublished - Jan 1 2019
Event6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018 - Ghent, Belgium
Duration: Oct 28 2018Oct 31 2018

Publication series

NameLife-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018

Conference

Conference6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018
CountryBelgium
CityGhent
Period10/28/1810/31/18

Fingerprint

Pipe
Neural networks
Cast iron pipe
Nodular iron
Cast iron
Electric power distribution
Managers
Water

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Safety, Risk, Reliability and Quality

Cite this

Kerwin, S., Garcia de Soto, B., & Adey, B. T. (2019). Performance comparison for pipe failure prediction using artificial neural networks. In D. M. Frangopol, R. Caspeele, & L. Taerwe (Eds.), Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018 (pp. 1337-1342). (Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018). CRC Press/Balkema.

Performance comparison for pipe failure prediction using artificial neural networks. / Kerwin, S.; Garcia de Soto, Borja; Adey, B. T.

Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018. ed. / Dan M. Frangopol; Robby Caspeele; Luc Taerwe. CRC Press/Balkema, 2019. p. 1337-1342 (Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kerwin, S, Garcia de Soto, B & Adey, BT 2019, Performance comparison for pipe failure prediction using artificial neural networks. in DM Frangopol, R Caspeele & L Taerwe (eds), Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018. Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018, CRC Press/Balkema, pp. 1337-1342, 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018, Ghent, Belgium, 10/28/18.
Kerwin S, Garcia de Soto B, Adey BT. Performance comparison for pipe failure prediction using artificial neural networks. In Frangopol DM, Caspeele R, Taerwe L, editors, Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018. CRC Press/Balkema. 2019. p. 1337-1342. (Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018).
Kerwin, S. ; Garcia de Soto, Borja ; Adey, B. T. / Performance comparison for pipe failure prediction using artificial neural networks. Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018. editor / Dan M. Frangopol ; Robby Caspeele ; Luc Taerwe. CRC Press/Balkema, 2019. pp. 1337-1342 (Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018).
@inproceedings{abb74fcee6c14e978db20ac23f6d2deb,
title = "Performance comparison for pipe failure prediction using artificial neural networks",
abstract = "Infrastructure managers must decide on the replacement timing of buried pipes in water distribution networks. These assets deteriorate resulting in failures and their associated consequences. In recent years, studies have investigated failure prediction models for pipes based on artificial neural networks (ANNs). These models are either generalized (i.e. trained with all pipe failures) or specialized (i.e. trained with certain pipe failures based on pipe material or failure history). It is currently unclear whether prediction accuracy is improved by developing several specialized ANNs compared to a single generalized one. To answer this question, four modelswere developed: a generalized model for cast iron (CI) and ductile iron (DI) pipes; a specialized model for CI pipes, a specialized model for CI pipes with no previous failures and a specialized model for CI pipes, which had previously failed. Overall, the study found minimal difference in performance between the generalized and specialized models.",
author = "S. Kerwin and {Garcia de Soto}, Borja and Adey, {B. T.}",
year = "2019",
month = "1",
day = "1",
language = "English (US)",
isbn = "9781138626331",
series = "Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018",
publisher = "CRC Press/Balkema",
pages = "1337--1342",
editor = "Frangopol, {Dan M.} and Robby Caspeele and Luc Taerwe",
booktitle = "Life-Cycle Analysis and Assessment in Civil Engineering",

}

TY - GEN

T1 - Performance comparison for pipe failure prediction using artificial neural networks

AU - Kerwin, S.

AU - Garcia de Soto, Borja

AU - Adey, B. T.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Infrastructure managers must decide on the replacement timing of buried pipes in water distribution networks. These assets deteriorate resulting in failures and their associated consequences. In recent years, studies have investigated failure prediction models for pipes based on artificial neural networks (ANNs). These models are either generalized (i.e. trained with all pipe failures) or specialized (i.e. trained with certain pipe failures based on pipe material or failure history). It is currently unclear whether prediction accuracy is improved by developing several specialized ANNs compared to a single generalized one. To answer this question, four modelswere developed: a generalized model for cast iron (CI) and ductile iron (DI) pipes; a specialized model for CI pipes, a specialized model for CI pipes with no previous failures and a specialized model for CI pipes, which had previously failed. Overall, the study found minimal difference in performance between the generalized and specialized models.

AB - Infrastructure managers must decide on the replacement timing of buried pipes in water distribution networks. These assets deteriorate resulting in failures and their associated consequences. In recent years, studies have investigated failure prediction models for pipes based on artificial neural networks (ANNs). These models are either generalized (i.e. trained with all pipe failures) or specialized (i.e. trained with certain pipe failures based on pipe material or failure history). It is currently unclear whether prediction accuracy is improved by developing several specialized ANNs compared to a single generalized one. To answer this question, four modelswere developed: a generalized model for cast iron (CI) and ductile iron (DI) pipes; a specialized model for CI pipes, a specialized model for CI pipes with no previous failures and a specialized model for CI pipes, which had previously failed. Overall, the study found minimal difference in performance between the generalized and specialized models.

UR - http://www.scopus.com/inward/record.url?scp=85063937688&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063937688&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85063937688

SN - 9781138626331

T3 - Life-Cycle Analysis and Assessment in Civil Engineering: Towards an Integrated Vision - Proceedings of the 6th International Symposium on Life-Cycle Civil Engineering, IALCCE 2018

SP - 1337

EP - 1342

BT - Life-Cycle Analysis and Assessment in Civil Engineering

A2 - Frangopol, Dan M.

A2 - Caspeele, Robby

A2 - Taerwe, Luc

PB - CRC Press/Balkema

ER -