Bayesian Network for Post-Earthquake Decision on Monitored Structures
S. Broglio, Armen Der Kiureghian · 2011
In this paper the problem of assessing the condition and making decisions for strategic structures, e.g. hospitals, police and fire stations, and highway/railway bridges, immediately after an earthquake is investigated. Nowadays it is possible to monitor such structures with placement of sensors, such as accelerometers, strain gages and seismographs. Due to their low cost and easy operation, it is possible to place a large number of such sensors in various parts of a structure or its surrounding ground. The question then arises as to how to post-process the large amount of data obtained from such sensors and from seismological sources that become available shortly after the earthquake. A Bayesian network (BN) framework for processing such evolving information is proposed herein. This probabilistic information, together with information on the costs associated with inspection and closure of the structure, is then used to formulate a decision problem regarding the optimal action to take in the aftermath of an earthquake. An example demonstrates the methodology for a typical building structure.