Flood Impacts on Road Transportation

Abstract

Flood disasters can penetrate every single aspect of human life and road transportation is no exception. However, flood impacts on road transportation is an area that has not been explored in detail in the past. The focus of this PhD study is on the performance assessment of a road network subject to flooding. In this work, several challenges were overcome with original ideas. The first was integrating the flood and the transport systems - both exhibiting strong temporal and spatial variations. This has been successfully achieved by implementing a novel methodology into a tool that modelled flood intensities output into a transport network constraint in a traffic model. The logic of the framework is intuitive – roads with shallow flood depth impose speed limitations, and roads with deep flood depth are closed for traffic. The developed tool enabled a quick and consistent technique to integrate the flood and the transport models in three different ways – static, semi-dynamic and dynamic. The static integration considers only one flood map to determine traffic conditions, whereas the semi-dynamic and the dynamic integrations use multiple maps to mimic the flood propagation in the traffic model. This thesis is the first to achieve semi-dynamic or dynamic integration of the two models. The second challenge was the assessment of the impacts. Intangible impacts such as travel delays propagating as knock-on effects can easily be misrepresented or even misunderstood. Employing a microscopic transport model allows for the assessment of direct effects and the knock-on consequences on individual drivers as well as the whole traffic system. Results in one case study suggest that the average travel time rose with 45% on average for 75% of the vehicles in the most affected hour of the simulation. The monetary value of traffic delays may not be as significant as the flood direct tangible damage, but flood impacts on road transportation may be more straightforward to alleviate if traffic authorities follow contingency plans to reduce traffic demand or mitigate potential interruptions of traffic supply. To analyse how potential interventions affect the transport system performance, three interventions were implemented into the model. The third challenge was the evaluation of the performance of a road transport system and the assessment of its resilience to flooding. Perusing this, a novel rationale to assess the resilience of a transport network has been developed. This method distinguishes reliability from resilience to define the nonlinear bounds of standard dry weather conditions, and any fluctuation beyond these bounds is defined as exceptional conditions. By separating reliability from resilience, the extent of both magnitude and duration is refined and contributes to better understanding of the system performance. This PhD thesis aspires to bridge the gap between flooding and traffic by providing a workable open source tool, which can be applied to other case studies and thus open the potential for further development in that area. As well as practical ideas, the theoretical contribution in assessing system resilience can be applied in other fields.