Management of Earthquake Risks using Condition Indicators

• Head of Project: Prof. Dr. A. Dazio
• Assistant: U. Yazgan
• Partner: Schweizerischer Nationalfonds
  Profs. Drs. M. Faber, H.R. Schalcher, A. Grün, Dr. J. Laue

Past earthquakes have shown that the poor seismic performance of structures and infrastructure can lead to large catastrophes. The effective management of earthquake risks is pivotal in mitigating such disastrous consequences of earthquakes on the society. The interdisciplinary project "Management of Earthquake Risks using Condition Indicators (MERCI)" involves six chairs of ETH and is aimed at developing an earthquake risk management framework where the observable characteristic descriptors (indicators) are rationally taken into account in the identification of optimal decisions regarding the risk. Within the framework of the MERCI project, the Earthquake Engineering and Structural Dynamics Group developed a set of modelling recommendations for simulating the seismic response of reinforced concrete structures as well as a novel post-earthquake damage assessment methodology.

In order to establish the modelling recommendations, a series of shaking table tests were numerically reproduced by adopting alternative modelling strategies. The accuracy of each modelling strategy in terms of estimating the characteristic response parameters was identified by comparing the simulated values against the measured ones. Both the maximum and the residual values of the response parameters were considered in the analyses. Moreover, the sensitivities of the predicted response parameters to the major modelling idealizations were investigated.

The developed post-earthquake damage assessment method allows the direct consideration of the observable damage and the measured residual deformations in the evaluation of the safety of damaged structures. The experienced maximum deformations are generally considered the determining factor in the assessment of seismic safety. The observable damage – i.e. indications of yielding or ultimate deformation being exceeded – that can be identified in a reconnaissance survey can help to improve the estimates of experienced maximum deformations. Furthermore, the measured residual deformations – i.e. the permanent deformation of the roof relative to the base and the rotations exhibited at the plastic hinge regions – are taken into account to further improve the estimates. Another novelty of the method is its direct consideration of the probable errors due to the inaccurateness of the numerical model that is utilized to predict the seismic response. The method was applied to a bridge model tested on a shaking table and promising results were obtained.