Attribute Selection and Bridge Performance Prediction Using Soft Computing Methods
| atmire.migration.oldid | 3448 | |
| dc.contributor.advisor | Cowe Falls, Lynne | |
| dc.contributor.author | Tagh Bostani, Maryam | |
| dc.date.accessioned | 2015-08-14T16:23:07Z | |
| dc.date.available | 2015-11-20T08:00:35Z | |
| dc.date.issued | 2015-08-14 | |
| dc.date.submitted | 2015 | en |
| dc.description.abstract | Bridge Management System (BMS) is a systematic decision-making process that uses engineering and economic principles to help transportation agencies manage the enormous number of bridges within a network. It seeks to deliver safe and sustainable bridges and minimize the Maintenance, Repair, and Rehabilitation (MR&R) costs during the life cycle of bridges. The Bridge Management framework starts with inspection and definition of the condition of bridges. Then, the future condition of bridges is predicted and the proper repair actions for a multi-year period are determined. Accurate prediction of the condition of bridges is vital in the effectiveness of BMS. Available bridge performance prediction methods have certain limitations in predicting the condition of bridges. This study addresses these limitations and introduces innovative, precise, and robust bridge performance prediction models. The attributes that are statistically significant to be predictors for bridge performance prediction are selected from the available database. Bridge inspection data are gathered repeatedly during time and hence are correlated. Generalized Estimating Equation (GEE) that is capable of handling the correlation within data is used to test the significance of each attribute. Backward Stepwise Elimination (BSE) and Genetic Algorithm (GA) are two procedures used to select the significant attributes according to GEE results. The selected attributes are then used to predict the condition of bridges using advanced Soft Computing methods. These biologically inspired methods mimic human-like intelligence to model the unknown target function by learning from available data. Local Linear Model Tree (LOLIMOT), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Support Vector Machines (SVM) are used for bridge performance prediction in two distinguished ways. In the first application, the condition of a specific bridge is predicted from available data on the other bridges in the network. In the second application, time series forecasting is used to predict the future condition of bridges from available data on their previous conditions. The models are verified using data from Alberta Transportation by predicting the condition of bridges in terms of Sufficiency Rating (SR). SR is an index that delivers a rational basis for priority planning of maintenance activities of bridges and allocating funds to the maintenance projects. | en_US |
| dc.identifier.citation | Tagh Bostani, M. (2015). Attribute Selection and Bridge Performance Prediction Using Soft Computing Methods (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/25066 | en_US |
| dc.identifier.doi | http://dx.doi.org/10.11575/PRISM/25066 | |
| dc.identifier.uri | http://hdl.handle.net/11023/2389 | |
| dc.language.iso | eng | |
| dc.publisher.faculty | Graduate Studies | |
| dc.publisher.institution | University of Calgary | en |
| dc.publisher.place | Calgary | en |
| dc.rights | University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. | |
| dc.subject | Engineering--Civil | |
| dc.subject.classification | Bridge Management | en_US |
| dc.subject.classification | Generalized Estimating Equation | en_US |
| dc.subject.classification | Local Linear Model Tree | en_US |
| dc.subject.classification | Adaptive Neuro Fuzzy Inference System | en_US |
| dc.subject.classification | Support Vector Machine | en_US |
| dc.subject.classification | Genetic Algorithm | en_US |
| dc.subject.classification | Sufficiency Rating | en_US |
| dc.title | Attribute Selection and Bridge Performance Prediction Using Soft Computing Methods | |
| dc.type | doctoral thesis | |
| thesis.degree.discipline | Civil Engineering | |
| thesis.degree.grantor | University of Calgary | |
| thesis.degree.name | Doctor of Philosophy (PhD) | |
| ucalgary.item.requestcopy | true |