Co-UDlabs. Building Collaborative Urban Drainage research lab communities.
Published August 24, 2022 | Version v1
Conference paper Open

Machine learning to improve understanding of sewer pipe failures

Creators

  • 1. University of Sheffield

Description

Paper presented in 10th International Conference on Sewer Processes and Networks

Highlights:

  • Machine Learning shows strong promise for interpreting linked pipe asset and failure data

  • Random Forests are able to predict if a pipe is at risk of causing a failure with high accuracy

  • Through predicting failure probability, pipes at risk can be identified for proactive inspection

Notes

This work is supported by the EU's H2020 research and innovation programme grant no. 101008626 and the UK's Engineering and Physical Sciences Research Council grant EP/S016813/1.

Files

SPN10_Kazemi-Machine_etal.pdf

Files (1.1 MB)

Name Size Download all
md5:b9c52b4662c3f342e404e5c95f0bcc44
1.1 MB Preview Download

Additional details

Funding

Co-UDlabs – Building Collaborative Urban Drainage research labs communities 101008626
European Commission