Seismic Risk Analysis of Transportation Networks Considering Socioeconomic Impacts

This paper targets the multidisciplinary area of seismic risk analysis of transportation networks. For this purpose, a new methodology and a library of probabilistic models are developed. Risk, in this context, denotes the probability of exceeding the sum of direct and indirect losses due to seismic events. Direct economic loss is due to the repair cost of the physical damage in the components of the network, such as bridges. The indirect loss is due to the reduced capacity of the components for traffic flow, which results in increased travel times in the network. This leads to socioeconomic loss that may even exceed the direct economic loss in severe earthquakes, yet it has been subject to limited studies. In this research, a reliability-based approach is adopted to carry out the risk analysis. For this purpose, many interacting probabilistic models including models for the earthquake hazard, infrastructure response, and direct and indirect consequences are needed to compute the losses. This study puts forward new models for probabilistic prediction of the damage, repair cost, and downtime of bridges, as one of the vulnerable components of transportation networks. To compute the amount of increase in the travel times of the damaged network, an artificial neural network is developed and trained by the data generated from analyzing the transportation network in the EMME/2 computer program. The methodology is showcased by a comprehensive application that features the transportation network of the City of Isfahan, the third largest city in Iran. The resulting loss curves indicate that the direct economic loss is, in general, more prominent that the socioeconomic loss. However, in small exceedance probabilities that represent very severe earthquake, the socioeconomic loss surpasses the direct loss.