15 Rapid Seismic Risk Assessment of Structures with Gaussian Process Regression 165 confident that we have the exact prediction of our responses and, for 97.1% of the time, this confidence predicts the exact damage state level or at most one damage state level apart. 15.4 Conclusion We propose to develop a model which utilizes limited inspection observations after earthquakes to predict building damage conditions based on Gaussian Process Regression. GPR model was able to predict responses with a satisfying accuracy of 92.7%. Moreover, considering the 95% confidence intervals for the predicted data, it was demonstrated that the model is 95% certain to predict the responses within at most one damage state apart, 97.1% of the time. Using this method, predictions can be made within a short period of time after the natural hazard for all the buildings in a vast area. References 1. Noh, H.Y., Lignos, D.G., Nair, K.K., Kiremidjian, A.S.: Development of fragility functions as a damage classification/prediction method for steel moment-resisting frames using a wavelet-based damage sensitive feature. Earthq. Eng. Struct. Dyn. 41(4), 681–696 (2012) 2. Kafali, C., Grigoriu, M.: Seismic fragility analysis: application to simple linear and nonlinear systems. Earthq. Eng. Struct. Dyn. 36(13), 1885–1900 (2007) 3. King, S., Kiremidjian, A., Pachakis, D., Sarabandi, P.: Application of empirical fragility functions from recent earthquakes. In: Proceedings of the 13th World Conference on Earthquake Engineering, Vancouver, BC, Canada. Paper (No. 2829) (2004) 4. Barbat, A.H., Pujades, L.G., Lantada, N.: Performance of buildings under earthquakes in Barcelona, Spain. Comput. Aided Civ. Inf. Eng. 21(8), 573–593 (2006) 5. Shibata, A.: Estimation of earthquake damage to urban systems. Struct. Control Health Monit. 13(1), 454–471 (2006) 6. Molas, G.L., Yamazaki, F.: Neural networks for quick earthquake damage estimation. Earthq. Eng. Struct. Dyn. 24(4), 505–516 (1995) 7. De Lautour, O.R., Omenzetter, P.: Prediction of seismic-induced structural damage using artificial neural networks. Eng. Struct. 31(2), 600–606 (2009) 8. Morfidis, K., Kostinakis, K.: Approaches to the rapid seismic damage prediction of r/c buildings using artificial neural networks. Eng. Struct. 165, 120–141 (2018) 9. Lagaros, N.D., Papadrakakis, M.: Neural network based prediction schemes of the non-linear seismic response of 3D buildings. Adv. Eng. Softw. 44(1), 92–115 (2012) 10. Seeger, M.: Gaussian processes for machine learning. Int. J. Neural Syst. 14(02), 69–106 (2004) 11. Ou, G.: Robust hybrid simulation with improved fidelity: theory, methodology, and implementation. Doctoral dissertation, Purdue University, West Lafayette (2016) 12. Ye, L., Ma, Q., Miao, Z., Guan, H., Zhuge, Y.: Numerical and comparative study of earthquake intensity indices in seismic analysis. Struct. Design Tall Spec. Build. 22(4), 362–381 (2013)
RkJQdWJsaXNoZXIy MTMzNzEzMQ==