An Advanced Statistical Approach to Data-Driven Earthquake Engineering

ABSTRACT

Decades-long experimental databases become accessible in global earthquake engineering community. Yet, complex interactions of a multitude of variables pose formidable challenges to data-driven research. We embarked upon developing an advanced statistical learning and prediction framework with the generalized additive model (GAM). We showed promising performance of GAM with applications to existing RC shear wall databases. Without any prejudice, GAM can predict structural responses accurately using raw databases, and also can identify salient attributes. This study addresses computational implementation and parallel processing, and all codes are made publicly available to promote data-driven research of earthquake engineering community.

KEYWORDS:

• Statistical Learning and Prediction
• Data-driven Earthquake Engineering
• Generalized Additive Model
• Reinforced Concrete Shear Wall

Acknowledgments

Special thanks are due to Professor John F. Hall for his productive guidance regarding nonlinear analysis methods, and also to Professor Sri Sritharan for valuable discussion on earthquake engineering experiments.

Additional information

Funding

This research is supported by the research funding of Department of Civil, Construction, and Environmental Engineering of Iowa State University. Generous research funding from Black & Veatch is appreciated. The simulations of this paper is partially supported by the HPC@ISU equipment at Iowa State University, some of which has been purchased through funding provided by NSF under MRI grant number [CNS 1229081] and CRI grant number [1205413].