Sohouenou, Philippe
(2022)
Assessing the role of network topology, demand variations and recovery strategies in road network resilience.
PhD thesis, University of Nottingham.
Abstract
Road networks are critical to society as they support people's daily mobility, the freight industry, and emergency services. However, a range of predictable and unpredictable events can affect road networks, disrupting traffic flows, connectivity, and, more generally, the functioning of society. With the increased interconnectivity and interdependency of the economic sectors, the need to manage this threat is more critical than ever. To this end, stakeholders need to understand the potential impacts of a multitude of predictable and unpredictable events. The present thesis aims at developing a framework to evaluate and understand the resilience (ability to sustain, resist and recover from perturbations) of road networks under a multitude of potentially unpredictable disruptions, and at assessing the role of different network design (e.g. network topology) and operation (e.g. travel-demand distribution) characteristics in road network resilience.
To this end, this thesis adopts a hazard-independent approach that considers all possible scenarios disrupting multiple links (more specifically up to a certain number of links). Novel indicators—including a robustness, unsatisfied-demand and resilience indicator measuring the demand-weighted-average increase in travel time in the disrupted network, the proportion of stranded travellers, and the speed of network-performance recovery, respectively—are developed and tested as part of this thesis. A link-criticality-assessment method based on multiple-link failures is also developed to identify the links that should be given priority for pre-event reinforcement and post-event restoration. To assess the effects of network size, topology, and demand distribution on network resilience, the thesis considers a variety of case studies, including artificial networks generated by a random road network model (developed as part of this thesis) and real network models derived from real-world maps. To assess the influence of demand variations, capacity constraints and congestion on network resilience, this thesis performs a resilience analysis of a network under several demand conditions. Finally, to assess the effects of recovery strategies on network resilience and characterize the optimal recovery strategy, this thesis performs a resilience analysis of a network considering all possible link-repair sequences.
This research should ultimately contribute to the incorporation of resilience considerations into transport planning and management standards, which currently give priority to transport efficiency—the efficient movement of vehicles through a transport network under normal conditions—rather than the movement of vehicles under disrupted conditions.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
