Keywords: Infrastructure maintenance, bridge lifetime, predictive maintenance
Our sister company, PEC measures the condition of bridges with physical surveys and advanced monitoring systems; it also modifies the condition of a bridge with its renovation activities. It would be desirable to integrate this information in a model that can evolve as the bridge ages. Once we have a model, we can use simulation to predict the future condition of the bridge. The model would have a physical component to describe and predict the condition of the bridge, but it would also have an economic component that would relate the physical changes of the bridge to its operating and maintenance costs.
The Life Cycle Cost (LCC) of a building is the cost of building, operating, maintaining and eventually demolishing the building. The predicted LCC will influence the design, specifications and maintenance of a building. We have developed a model of the LCC that is specifically tailored to bridges and is designed to support planners making investment decisions. The model predicts how a bridge’s components will deteriorate; when and where intervention is required and the costs and benefits of the intervention.
LCC Analysis may be applied at several levels of detail: our model works at the level of bridge components, such as bearings. Components can be identified, inspected and can normally be repaired without affecting the rest of the bridge. Some of these components, such as bearings and expansion joints, may not be present in smaller bridges. Even so, our model is sufficiently flexible to be able to cope with a wide range of bridges.
The lifecycle of a component: installation, various stages of deterioration, renewal and termination is modelled as a series of states, or cohorts. The initial state of the model can be informed by a bridge survey or monitoring. As the component ages, it progresses from state to state with a given probability. This probability of a transition from one state to another, for a given component type, is estimated from historical databases. With this model, it is possible to rigorously predict the likely future condition of the bridge, to simulate various scenarios and the results of different actions.
The ideas behind our LCC model are described more fully in [1].
[1] Petschacher, M. “Probabilistic ageing model for infrastructure buildings”, Structure and Infrastructure Engineering 5(4): 295-300, August 2009