Framework for probabilistic connectivity estimation of bridge networks under seismic and flood hazards

Abstract

As observed in recent disasters due to earthquakes and floods (e.g. 2011 Great East Japan earthquake and 2020 Kyushu floods), since the transportation networks including bridges play a crucial role in the evacuation of affected people and the transportation of emergency goods and materials, the functionality of the network after a disaster must be investigated. A significant amount of research efforts has shifted the focus from the investigation of the performance of individual components of the infrastructure to that of entire distributed civil infrastructure systems and networks. Therefore, to develop disaster mitigation strategies in both earthquake and flood-prone regions, it is essential to evaluate the deterioration of bridge network functionality considering the damage to individual bridges subjected to ground motion and flood actions. This paper presents a novel framework for probabilistic connectivity assessment of bridge networks under seismic and flood hazards. Although a retrofitting prioritization as a mitigation strategy for bridges is the key to minimize the potential losses, it should be noted that retrofitting strategies to maximize the connectivity of the road network under seismic hazard will not necessarily result in maximizing the connectivity of the road network under flood hazard. Effect of seismic and flood hazards and bridge fragility on the road network connectivity should be taken into consideration for identifying the dominant hazard and bridges requiring retrofitting.

Peak ground velocity and annual peak discharge are selected as the intensity measures for seismic and flood hazards, respectively. Both hazards are quantified considering the difference of soil condition and topography among bridge locations. Conditional failure probability is estimated according to the possible multiple failure modes of bridges under seismic and flood hazards. With the information on hazard and vulnerability of individual bridges, Monte-Carlo simulation is performed to calculate the probabilistic connectivity of a bridge network. In this study, connectivity is defined as the probability that a path does not exist between origin-destination (OD) pairs after a disaster event.

The proposed framework is applied to hypothetical bridge networks along the Kuma River in Kumamoto prefecture, Japan. Kumamoto prefecture has had several big earthquakes and devastating floods. After conducting the hazard analysis and vulnerability assessments of the bridge structures, the connectivity of bridge networks considering the spatial correlations associated with the hazard intensities and bridge damages is evaluated, and optimal disaster mitigation strategy is presented and discussed.