Unification of Multi-Source and Multi-Fidelity Fragility Functions

Published

April 4, 2026

Author(s)

Mohammadamin Hariri Ardebili, Siamak Sattar

Abstract

The derivation of fragility functions through nonlinear dynamic analysis is a domain that has garnered significant attention historically. The majority of prior research efforts have been focused on the ground motion record-to-record variability and the incorporation of material randomness within numerical simulations. However, there remains an underexplored area concerning the uncertainties associated with the numerical modeling of structural physics. Diverse numerical models, ranging from low to high fidelity, necessitate varying degrees of material and modeling parameters, leading to disparate fragility outcomes. This divergence underscores the imperative need for methodologies that combine these varying functions into a cohesive framework for risk evaluation. This paper introduces a framework that offers varied practical methodologies for the formulation of fragility functions, which encapsulate the triad of uncertainties: record-to-record variability in ground motion, material property randomness, and modeling strategy selections. Utilizing a case study of a reinforced concrete bridge pier, supplemented by shake table experimental results and the findings from a blind prediction contest, the study evaluates a suite of approaches for unifying multiple fragility functions obtained from different uncertainty sources and model fidelities. These approaches include decision tree selection, uniform/non-uniform weighting, mixed engineering demand parameter integration, and variance–bias decomposition to synthesize a unified fragility representation. The findings highlight the substantial variability in outcomes based on data accessibility, underlying assumptions in the combination process, and the fidelity of the employed models.

Citation

Reliability Engineering & System Safety

Pub Type

Journals

Download Paper

Local Download

Citation

Hariri Ardebili, M. and Sattar, S. (2026), Unification of Multi-Source and Multi-Fidelity Fragility Functions, Reliability Engineering & System Safety, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=960482 (Accessed May 15, 2026)

Additional citation formats