Infrastructure Condition Assessment and Prediction under Variable Traffic Demand and Management Scenarios

dc.contributor.authorAbi Aad, Mirlaen
dc.contributor.committeechairAbbas, Montasir M.en
dc.contributor.committeememberTrani, Antonio A.en
dc.contributor.committeememberWang, Linbingen
dc.contributor.committeememberZhang, Wenwenen
dc.contributor.departmentCivil and Environmental Engineeringen
dc.date.accessioned2022-11-09T09:00:11Zen
dc.date.available2022-11-09T09:00:11Zen
dc.date.issued2022-11-08en
dc.description.abstractDepartments of Transportation (DOTs) are responsible for keeping their road network in a state of good repair while also aiming to reduce congestion through the implementation of different traffic control and demand management strategies. These strategies can result in changes in traffic volume distributions, which in turn affect the level of pavement deterioration due to traffic loading. To address this issue, this dissertation introduces an integrated simulation-optimization framework that accounts for the combined effects of pavement conditions and traffic management decision-making strategies. The research focuses on exploring the range of possible performance outcomes resulting from this integrated modeling approach. The research also applied the developed framework to a particular traffic demand management strategy and assessed the impact of dynamic tolls around the specific site of I-66 inside the beltway. The integrated traffic-management/pavement-treatment framework was applied to address both the operational and pavement performance of the network. Aimsun hybrid macro/meso dynamic user equilibrium experiments were used to simulate the network with a modified cost function taking care of the dynamic pricing along the I-66 tolled facility. Furthermore, the framework was expanded to include the development of a systematic and comprehensive methodology to optimize the allocation of networkwide pavement treatment work zones over space and time. The proposed methodology also contributed to the development of a surrogate function that reduces the optimization computation burden so that researchers would be able to conduct work zone allocation optimization without having to run expensive simulation work. Finally, in this dissertation, a user-friendly decision-support tool was developed to assist in the pavement treatment and project selection planning process. We use machine learning models to encapsulate the simulation optimization process.en
dc.description.abstractgeneralDepartments of Transportation (DOTs) are responsible for keeping their road network in a state of good repair. Improvement in pavement rehabilitation plans can lead to savings in the order of tens of millions of dollars. Pavement rehabilitation plans result in work zone schedules on the transportation network. Limited roadway capacities due to work zones affect traffic assignments and routing on the roads, which impacts the selection of optimal operation strategies to manage the resulting traffic. On the other hand, the choice of any particular operation and routing strategy will result in different distributions of traffic volumes on the roads and affect the pavement deterioration levels due to traffic loading, leading to other optimal rehabilitation plans and corresponding work zones. While there have been several research efforts on infrastructure condition assessment and other research efforts on traffic control and demand management strategies, there is a wide gap in the nexus of the two fields. To address this issue, this dissertation introduces an integrated simulation-optimization framework that accounts for the combined effects of pavement conditions and traffic management decision-making strategies. The research focuses on exploring the range of possible performance outcomes resulting from this integrated modeling approach. The research also applied the developed framework to a particular traffic demand management strategy and assessed the impact of dynamic tolls around the specific site of I-66 inside the beltway. The integrated traffic-management/pavement-treatment framework was applied to address both the operational and pavement performance of the network. Furthermore, the framework was expanded to include the development of a systematic and comprehensive methodology to optimize the allocation of networkwide pavement treatment work zones over space and time. The proposed methodology also contributed to the development of a surrogate function that reduces the optimization computation burden so that researchers would be able to conduct work zone allocation optimization without having to run expensive simulation work. Finally, in this dissertation, a user-friendly decision-support tool was developed to assist in the pavement treatment and project selection planning process. We use machine learning models to encapsulate the simulation optimization process.en
dc.description.degreeDoctor of Philosophyen
dc.format.mediumETDen
dc.identifier.othervt_gsexam:35725en
dc.identifier.urihttp://hdl.handle.net/10919/112543en
dc.language.isoenen
dc.publisherVirginia Techen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectPavement Conditionsen
dc.subjectTraffic Assignmenten
dc.subjectModelingen
dc.subjectPavement Treatmenten
dc.subjectOptimizationen
dc.subjectSimulationen
dc.subjectTraffic Demand Managementen
dc.subjectWork Zoneen
dc.subjectMachine Learningen
dc.titleInfrastructure Condition Assessment and Prediction under Variable Traffic Demand and Management Scenariosen
dc.typeDissertationen
thesis.degree.disciplineCivil Engineeringen
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen
thesis.degree.leveldoctoralen
thesis.degree.nameDoctor of Philosophyen

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Abi_Aad_M_D_2022.pdf
Size:
48.58 MB
Format:
Adobe Portable Document Format