General Sewer Deterioration Model Using Random Forest

Bolette Dybkjær Hansen, Thomas B. Moeslund, David Getreuer Jensen, Jamshid Tamouk, Mads Uggerby, Søren Højmark Rasmussen

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

Resumé

Collapse of sewers can induce significant damage
to roads and buildings, resulting in large economical costs.
Therefore, utilities wish to repair or replace the sewers before
they collapse. In order to investigate if a sewer needs maintenance
or replacement it can be inspected with Closed Circuit Television
(CCTV), but as CCTV inspection is very expensive, and hence
only a small percentage of the sewers are inspected. This underlines
the importance of choosing the correct sewers for inspection
and have resulted in development of several deterioration models.
However, the best performing existing models are tailored to
individual cities and need to be calibrated in order to be
generalized to new areas. As the cost for collecting a data set
for calibration is high, the utilities could benefit from a sewer
deterioration model which generalizes across location.
This paper presents a deterioration model based on Random
Forest, which is trained on data from 35 utilities spread across
the country of Denmark. The model was able to predict the
sewer condition with a specificity at 0.80 and a sensitivity at 0.76,
which is comparable to the best existing models. This shows that
it is possible to make a deterioration model which generalizes
across data from different regions, sewers and utilities. This is a
significant improvement compared to the current situation where
models need to be learned for each new set of data.
OriginalsprogDansk
TitelIEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES)
Antal sider8
Publikationsdato2020
StatusUdgivet - 2020

Emneord

  • sewers
  • deterioration
  • machine learning
  • maintenance
  • data science
  • ageing
  • infrastructure

Citer dette

Hansen, B. D., Moeslund, T. B., Jensen, D. G., Tamouk, J., Uggerby, M., & Rasmussen, S. H. (2020). General Sewer Deterioration Model Using Random Forest. I IEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES)
Hansen, Bolette Dybkjær ; Moeslund, Thomas B. ; Jensen, David Getreuer ; Tamouk, Jamshid ; Uggerby, Mads ; Rasmussen, Søren Højmark. / General Sewer Deterioration Model Using Random Forest. IEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES). 2020.
@inproceedings{6b0645b986554f63891e94095da842b9,
title = "General Sewer Deterioration Model Using Random Forest",
abstract = "Collapse of sewers can induce significant damageto roads and buildings, resulting in large economical costs.Therefore, utilities wish to repair or replace the sewers beforethey collapse. In order to investigate if a sewer needs maintenanceor replacement it can be inspected with Closed Circuit Television(CCTV), but as CCTV inspection is very expensive, and henceonly a small percentage of the sewers are inspected. This underlinesthe importance of choosing the correct sewers for inspectionand have resulted in development of several deterioration models.However, the best performing existing models are tailored toindividual cities and need to be calibrated in order to begeneralized to new areas. As the cost for collecting a data setfor calibration is high, the utilities could benefit from a sewerdeterioration model which generalizes across location.This paper presents a deterioration model based on RandomForest, which is trained on data from 35 utilities spread acrossthe country of Denmark. The model was able to predict thesewer condition with a specificity at 0.80 and a sensitivity at 0.76,which is comparable to the best existing models. This shows thatit is possible to make a deterioration model which generalizesacross data from different regions, sewers and utilities. This is asignificant improvement compared to the current situation wheremodels need to be learned for each new set of data.",
keywords = "sewers, deterioration, machine learning, maintenance, data science, ageing, infrastructure",
author = "Hansen, {Bolette Dybkj{\ae}r} and Moeslund, {Thomas B.} and Jensen, {David Getreuer} and Jamshid Tamouk and Mads Uggerby and Rasmussen, {S{\o}ren H{\o}jmark}",
year = "2020",
language = "Dansk",
booktitle = "IEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES)",

}

Hansen, BD, Moeslund, TB, Jensen, DG, Tamouk, J, Uggerby, M & Rasmussen, SH 2020, General Sewer Deterioration Model Using Random Forest. i IEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES).

General Sewer Deterioration Model Using Random Forest. / Hansen, Bolette Dybkjær; Moeslund, Thomas B.; Jensen, David Getreuer; Tamouk, Jamshid; Uggerby, Mads; Rasmussen, Søren Højmark.

IEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES). 2020.

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - General Sewer Deterioration Model Using Random Forest

AU - Hansen, Bolette Dybkjær

AU - Moeslund, Thomas B.

AU - Jensen, David Getreuer

AU - Tamouk, Jamshid

AU - Uggerby, Mads

AU - Rasmussen, Søren Højmark

PY - 2020

Y1 - 2020

N2 - Collapse of sewers can induce significant damageto roads and buildings, resulting in large economical costs.Therefore, utilities wish to repair or replace the sewers beforethey collapse. In order to investigate if a sewer needs maintenanceor replacement it can be inspected with Closed Circuit Television(CCTV), but as CCTV inspection is very expensive, and henceonly a small percentage of the sewers are inspected. This underlinesthe importance of choosing the correct sewers for inspectionand have resulted in development of several deterioration models.However, the best performing existing models are tailored toindividual cities and need to be calibrated in order to begeneralized to new areas. As the cost for collecting a data setfor calibration is high, the utilities could benefit from a sewerdeterioration model which generalizes across location.This paper presents a deterioration model based on RandomForest, which is trained on data from 35 utilities spread acrossthe country of Denmark. The model was able to predict thesewer condition with a specificity at 0.80 and a sensitivity at 0.76,which is comparable to the best existing models. This shows thatit is possible to make a deterioration model which generalizesacross data from different regions, sewers and utilities. This is asignificant improvement compared to the current situation wheremodels need to be learned for each new set of data.

AB - Collapse of sewers can induce significant damageto roads and buildings, resulting in large economical costs.Therefore, utilities wish to repair or replace the sewers beforethey collapse. In order to investigate if a sewer needs maintenanceor replacement it can be inspected with Closed Circuit Television(CCTV), but as CCTV inspection is very expensive, and henceonly a small percentage of the sewers are inspected. This underlinesthe importance of choosing the correct sewers for inspectionand have resulted in development of several deterioration models.However, the best performing existing models are tailored toindividual cities and need to be calibrated in order to begeneralized to new areas. As the cost for collecting a data setfor calibration is high, the utilities could benefit from a sewerdeterioration model which generalizes across location.This paper presents a deterioration model based on RandomForest, which is trained on data from 35 utilities spread acrossthe country of Denmark. The model was able to predict thesewer condition with a specificity at 0.80 and a sensitivity at 0.76,which is comparable to the best existing models. This shows thatit is possible to make a deterioration model which generalizesacross data from different regions, sewers and utilities. This is asignificant improvement compared to the current situation wheremodels need to be learned for each new set of data.

KW - sewers

KW - deterioration

KW - machine learning

KW - maintenance

KW - data science

KW - ageing

KW - infrastructure

M3 - Konferenceartikel i proceeding

BT - IEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES)

ER -

Hansen BD, Moeslund TB, Jensen DG, Tamouk J, Uggerby M, Rasmussen SH. General Sewer Deterioration Model Using Random Forest. I IEEE Symposium on Computational Intelligence for Engineering Solutions (IEEE CIES). 2020