A methodology for prioritizing simulation-based stress tests for transportation systems


METADATA ONLY
Loading...

Date

2024-05

Publication Type

Other Conference Item

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

Transportation systems are integral to the economic development and prosperity of communities. Their service, however, is affected by disruptive events causing economic and socio-economic consequences. To ensure these consequences are acceptably minimal under various disruptive scenarios, stress testing, as a diagnostic approach, has shown promise. A stress test, in this context, represents a stressed situation, where at least one variable describing the system, e.g., hazard intensity or performance of assets, is significantly worse than expected. Nasrazadani et al. (Nasrazadani et al. 2024) proposed a method to define and conduct simulation-based stress tests on transportation systems. Their approach entails an initial reference risk assessment, where the system is in its baseline condition, followed by conducting stress tests, which represent stressed situations. Although simulation-based stress tests offer detailed insights into system's resilience, their computational demand makes it impractical to conduct all potential tests in real applications. The challenge remains in determining which stress tests, among numerous possibilities, are more critical to be conducted. To address this gap, this research proposes a novel computationally efficient method to prioritize candidate stress tests. The proposed methodology features a novel implementation of importance sampling approach that utilizes only the results of the reference assessment to identify which stress tests, if were to be conducted explicitly using simulations, would potentially have a higher impact on elevating the risks. For each stress test, a resampled subset of the results of the reference assessment is selected such that it would realize the specific conditions of that stress test, and thus its potential impact on risks. Those stress tests that are shown to lead to higher increase in risks in this approach are expected to do so as well if conducted explicitly, and hence deemed more critical. The methodology is applied to a road network in Switzerland subjected to extreme scenarios of rainfall flooding and landslides. The proposed methodology allows infrastructure managers to conduct a screening analysis of candidate stress tests, identify the critical ones without explicitly conducting them, and allocate resources to conduct only the most critical ones. This reduces the number of potential scenarios while maximizing insights into the system’s resilience. Additionally, it identifies which stress tests, if improved by taking risk-reducing measures, can contribute to higher improvement in system’s resilience.

Permanent link

Publication status

published

External links

Editor

Book title

Book of Abstracts EMI/PMC 2024

Journal / series

Volume

Pages / Article No.

Publisher

American Society of Civil Engineers

Event

Engineering Mechanics Institute Conference and Probabilistic Mechanics and Reliability Conference (EMI/PMC 2024)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Organisational unit

03859 - Adey, Bryan T. / Adey, Bryan T. check_circle

Notes

Presentation

Funding

Related publications and datasets