ABSTRACT
Determining the intervention strategies for bridges should consider all possible ways that bridges can evolve over time. This means all possible failures at all possible times should be considered as well as how the infrastructure will be restored, and the effect of the restoration interventions on future behavior of the infrastructure. Infrastructure failures include those that occur due to gradual processes, e.g. the chloride induced corrosion that leads to an expansive rust product in a concrete bridge, and those that occur due to sudden failure processes, e.g. the washing away of a bridge due to high flood waters. The former occurs in much literature that is focused on the determination of optimal interventions strategies. The latter is normally dealt with under the guise of risk analysis. From a bridge management perspective, the former is acceptable when the probability of failure due to sudden processes is negligible, and the latter is acceptable when the probability of failure is so high so that interventions that might be required due to gradual processes can be neglected. This, however, still leaves a large number of possibilities where both should be considered. Recent research has shown that it is possible to determine optimal intervention strategies, taking into consideration both gradual deterioration processes and sudden changes in environmental conditions, at a high level of abstraction. A high level of abstraction is normally, however, something that might be satisfactory for making high level budget estimates but is lacking when it comes to estimating the specific interventions that should be executed on a bridge. With this in mind, this paper proposes a methodology to determine of optimal intervention strategies taking into consideration both the effects of sudden changes in environmental conditions and the effects of gradual deterioration processes. At this scope, fault trees are used to estimate the probability and impact of failures in each unit of time taking into consideration the condition of the objects, and the Monte Carlo method is used to simulate bridge behavior over time.
