ATTENTION: The works hosted here are being migrated to a new repository that will consolidate resources, improve discoverability, and better show UTA's research impact on the global community. We will update authors as the migration progresses. Please see MavMatrix for more information.
Show simple item record
dc.creator | Loganathan, Karthikeyan | |
dc.date.accessioned | 2021-09-14T16:18:08Z | |
dc.date.available | 2021-09-14T16:18:08Z | |
dc.date.created | 2021-08 | |
dc.date.issued | 2021-08-26 | |
dc.date.submitted | August 2021 | |
dc.identifier.uri | http://hdl.handle.net/10106/30023 | |
dc.description.abstract | Underground utilities and wastewater collection systems deteriorate over time demanding utility owners to involve in continuous revisions and development of their asset management frameworks to maintain the functionality of their assets. In any asset management framework, inspection of an asset and respective condition assessment plays a vital role in successful operation and maintenance of systems. In the United States, closed-circuit television (CCTV) is the commonly used device for inspecting the inner environment of sewer pipes, which considering the large length of pipe inventory in a city, is a relatively expensive and time-consuming process. Therefore, inspection of every individual sanitary sewer pipe segment is not feasible in a short time period for any municipality owing to their large inventory of these pipes. However, sanitary sewer pipe segments in need of repair or a maintenance activity can be prioritized in advance for inspection based on their historical performance. Therefore, the primary objective of this dissertation is to develop a sanitary sewer pipe condition prediction model. Data collected from City of Fort Worth, Texas, is utilized in model development. Various supervised machine learning algorithms such as logistic regression (LR), k-nearest neighbors (k-NN) and random forests (RF) are employed. Numerous evaluation metrics such as precision, recall, F1-score and area under curve (AUC) are estimated to compare the performance of developed models. Resulted F1-score for the RF model is 0.94 while LR and k-NN models resulted 0.83 and 0.44, respectively. The results show that random forests model performed better than both LR and k-NN models. As a secondary objective of this dissertation, a decision support tool was developed for asset managers to utilize above models during inspection phase to estimate condition of their sanitary sewers for identification of critical sewers in need of immediate attention. | |
dc.format.mimetype | application/pdf | |
dc.language.iso | en_US | |
dc.subject | Condition Assessment | |
dc.subject | Inspection | |
dc.title | Development of a Model to Prioritize Inspection and Condition Assessment of Gravity Sanitary Sewer Systems | |
dc.type | Thesis | |
dc.degree.department | Civil Engineering | |
dc.degree.name | Doctor of Philosophy in Civil Engineering | |
dc.date.updated | 2021-09-14T16:18:09Z | |
thesis.degree.department | Civil Engineering | |
thesis.degree.grantor | The University of Texas at Arlington | |
thesis.degree.level | Doctoral | |
thesis.degree.name | Doctor of Philosophy in Civil Engineering | |
dc.type.material | text | |
dc.creator.orcid | 0000-0002-4508-7045 | |
Files in this item
- Name:
- LOGANATHAN-DISSERTATION-2021.pdf
- Size:
- 3.865Mb
- Format:
- PDF
This item appears in the following Collection(s)
Show simple item record