Probabilistic bridge network life-cycle connectivity assessment and optimization

A methodology and a numerical tool for the optimal scheduling of the maintenance interventions on a bridge network considering connectivity and maintenance cost are presented. The main objective of the optimization is the network connectivity. In this way, the probability that, even after an extreme event, it is possible to reach all the areas of the region using the highway system is maximized. The connectivity is assessed and predicted using techniques from graph theory and transportation engineering. A second objective of the optimization is the minimization of the maintenance cost along the life-cycle of the network. This optimization problem with two conflicting criteria is solved using multi-objective genetic algorithms. The large amount of uncertainty involved in the problem at hand (i.e. time-dependent reliability of the bridges, effects of the maintenance actions, correlation among the bridge in/out of service states) is accounted for using stochastic models, Latin hypercube sampling, and random field simulation.