Defect based deterioration model for sewer pipelines using bayesian belief networks
Author | Elmasry, Mohamed |
Author | Hawari, Alaa |
Author | Zayed, Tarek |
Available date | 2023-05-23T09:39:16Z |
Publication Date | 2017 |
Publication Name | Canadian Journal of Civil Engineering |
Resource | Scopus |
Abstract | A defect based deterioration model to determine the condition ratings in a probabilistic manner for sewer pipelines is presented in this paper. Bayesian belief network (BBN) is used to develop a static model using probabilities of occurrences, and conditional probabilities from observations of existing sewage network. Time dimension is introduced to the developed BBN model by using logistic regression as temporal links required to construct a dynamic Bayesian belief network (DBN). The accuracy of the model’s prediction is examined using actual data where the mean absolute error and root mean square error for the BBN model resulted in values of 0.67, 1.06, 0.56 and 1.05, 1.60, 0.95 for structural, operational, and overall conditions, respectively. As for the DBN model, values achieved for the year at which a pipeline would reach a certain condition state were close to the actual values from the validation dataset. |
Sponsor | This publication was made possible by NPRP grant # (NPRP6-357-2-150) from the Qatar National Research Fund (a member of The Qatar Foundation). The statements made herein are solely the responsibility of the authors. Also the authors would like to thank the public works authority of Qatar (ASHGAL) for their support in the data collection. |
Language | en |
Publisher | Canadian Science Publishing |
Subject | Bayesian belief network Deterioration model Dynamic Bayesian network Monte Carlo simulation Multinomial logistic regression Sewer pipelines defects |
Type | Article |
Pagination | 675-690 |
Issue Number | 9 |
Volume Number | 44 |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Civil and Environmental Engineering [778 items ]