Bayesian Surrogates for Risk-Aware Pre-Assessment of Aging Bridge Portfolios

2025-09-29

AI Summary

In resource allocation for aging bridge populations, a fundamental trade-off exists between high-fidelity simulation—computationally prohibitive at scale—and low-cost methods that lack controllable risk assessment. Method: This paper proposes a Bayesian neural network (BNN) surrogate model calibrated for epistemic uncertainty. Trained on a large-scale dataset generated via parametric modeling and nonlinear finite element analysis, it rapidly predicts code compliance factors for bridges, enabling risk-aware preliminary evaluation and prioritization. Contribution/Results: Unlike conventional approaches, the BNN preserves rigorous uncertainty quantification while drastically improving computational efficiency, thereby supporting decision optimization across full combinatorial design spaces. Case studies demonstrate that the framework accurately identifies structures requiring high-fidelity analysis, reducing redundant simulations and unnecessary maintenance interventions—ultimately lowering lifecycle costs and carbon emissions.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Aging infrastructure portfolios pose a critical resource allocation challenge: deciding which structures require intervention and which can safely remain in service. Structural assessments must balance the trade-off between cheaper, conservative analysis methods and accurate but costly simulations that do not scale portfolio-wide. We propose Bayesian neural network (BNN) surrogates for rapid structural pre-assessment of worldwide common bridge types, such as reinforced concrete frame bridges. Trained on a large-scale database of non-linear finite element analyses generated via a parametric pipeline and developed based on the Swiss Federal Railway's bridge portfolio, the models accurately and efficiently estimate high-fidelity structural analysis results by predicting code compliance factors with calibrated epistemic uncertainty. Our BNN surrogate enables fast, uncertainty-aware triage: flagging likely critical structures and providing guidance where refined analysis is pertinent. We demonstrate the framework's effectiveness in a real-world case study of a railway underpass, showing its potential to significantly reduce costs and emissions by avoiding unnecessary analyses and physical interventions across entire infrastructure portfolios.

Problem

• Balancing cost-effective conservative analysis with accurate simulations for aging bridges
• Providing rapid structural pre-assessment with calibrated uncertainty estimates for bridge portfolios
• Reducing unnecessary analyses and interventions across entire infrastructure systems

Innovation

• Bayesian neural network surrogates for bridge assessment
• Trained on parametric finite element analysis database
• Predicts code compliance with calibrated uncertainty