Surrogate learning for scoured bridges for capacity prediction with climate-scenario deep uncertainties

Waterway bridges are among the most vulnerable components of transportation infrastructure due to scour—the erosion of soil around foundations caused by flowing water—a risk expected to grow under climate-driven increases in flood intensity. Traditional fragility models often assume a fixed structural capacity, overlooking how scour alters soil–foundation–structure interaction (SFSI) and degrades performance.This study develops a computationally efficient surrogate-learning framework, trained on nonlinear pushover analyses, to predict yield-level transverse capacities—base shear, deck displacement, base moment, and column rotation—collectively expressed as a capacity tuple. To incorporate deep uncertainty arising from future IPCC climate-scenario trajectories, we employ a credal-set approach that yields upper and lower bounds on capacity predictions. The resulting method enables rapid, risk-informed evaluation of scour-critical bridges, supporting practical decision-making under uncertain future hydraulic hazards.

Name thesis:degree_name

M.S. (Master of Science)

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Civil Engineering (UMKC)

Grantor

University of Missouri--Kansas City

Year dc:date.issued

2025

Author dc:creator

• Qadir, Ilham

Advisor dc:contributor.advisor

• Chen, ZhiQiang

Handle dc:identifier.uri

https://hdl.handle.net/10355/109201

OAI identifier oai:identifier

oai:mospace.umsystem.edu:10355/109201