Stripe‐based fragility analysis of multispan concrete bridge classes using machine learning techniques

Abstract

Summary A framework for the generation of bridge‐specific fragility curves utilizing the capabilities of machine learning and stripe‐based approach is presented in this paper. The proposed methodology using random forests helps to generate or update fragility curves for a new set of input parameters with less computational effort and expensive resimulation. The methodology does not place any assumptions on the demand model of various components and helps to identify the relative importance of each uncertain variable in their seismic demand model. The methodology is demonstrated through the case study of a multispan concrete bridge class in California. Geometric, material, and structural uncertainties are accounted for in the generation of bridge numerical models and their fragility curves. It is also noted that the traditional lognormality assumption on the demand model leads to unrealistic fragility estimates. Fragility results obtained by the proposed methodology can be deployed in a risk assessment platform such as HAZUS for regional loss estimation.

The promise of implementing machine learning in earthquake engineering: A state-of-the-art review

“…Xie and DesRoches (2019) tested stepwise and LASSO regression models for probabilistic seismic demand analysis of California highway bridges. Other than RSMs, multi-predictor PSDMs have been developed using ANN (Calabrese and Lai, 2013; Lagaros et al, 2009; Lagaros and Fragiadakis, 2007; Liu and Zhang, 2018; Mitropoulou and Papadrakakis, 2011; Pang et al, 2014; Wang et al, 2018), bootstrapped ANN (Ferrario et al, 2017), SVM (Ghosh et al, 2018; Hariri-Ardebili and Pourkamali-Anaraki, 2018; Huang et al, 2017; Mahmoudi and Chouinard, 2016), kriging metamodeling (Gidaris et al, 2015), GLM (Xie et al, 2019a), MARS (Kameshwar and Padgett, 2014), K-nearest neighbor (Hariri-Ardebili and Pourkamali-Anaraki, 2018), naïve Bayes classifier (Hariri-Ardebili and Pourkamali-Anaraki, 2018), high-dimensional model representation (Sahu et al, 2019), and RF (Mangalathu and Jeon, 2019b).…”

Metamodel-Based Seismic Fragility Analysis of Concrete Gravity Dams

“…This combination is justified by the significant randomness that characterizes not only the earthquake excitation but also the structural system itself (e.g., stochastic variations in the material properties, degradation due to aging, and temperature fluctuation, etc.). Surrogate modeling techniques within a seismic fragility framework have found recent applications for the safety assessment of buildings and bridges, among other structures (Mangalathu and Jeon 2018;Sichani et al 2017;Kameshwar and Padgett 2014;Ghosh et al 2013;Seo and Linzell 2013;Seo et al 2012). Even though many of these studies considered several seismic intensity measures (IMs) and model parameters (MPs) for building the metamodels to predict the response of the structure, most of them do not clearly depict the influence of all the considered parameters in the form of multivariate fragility functions from the respective metamodels.…”

Vibration characteristic analyses of medium-and small-span girder bridge groups in highway systems based on machine learning models

“…Sun et al (2019) utilized ridge regressions and support vector machines to analyze the influences of eight specific factors on the peak drift ratios and peak floor accelerations of four typical buildings and to effectively capture the nonlinear responses of high-rise buildings. Mangalathu et al (2019) and Mangalathu and Jeon (2019), used random forest models to establish the seismic fragility curves for California bridges, and then proposed a method for rapid damage state assessments using ML techniques. Jia et al (2020) sorted the importance levels of the specific characteristics and identified the seismic damages to the bridges following the Wenchuan earthquake according to the ML models.…”