Machine Learning-Based Risk Analysis for Infrastructure Vulnerable to Flood Hazards
Publication: Construction Research Congress 2024
ABSTRACT
Flood hazards have affected millions of people through damage to infrastructure. Flood hazards in the Great Lakes area in recent years have caused a large area of the infrastructure system in Indiana and Illinois to be paralyzed due to the impact of the high flood level on infrastructure. To assess the risk of floods for certain types of infrastructure, we develop a machine learning-based risk analysis model to generate predictions of risk scores based on several indicators from a public dataset. Different infrastructure types such as bridges, culverts, pump stations, and roads were considered. We predicted risk scores to represent the flood risk from multiple components, such as exposure and criticality to flood hazards. The correlation between certain indicators and risk scores was examined to determine the critical variables used as input for the risk analysis model. The Gaussian process regression (GPR) algorithm was applied to train and test the dataset. From the analysis results, we find the GPR model provides accurate prediction. Therefore, the machine learning-based risk analysis is a valid approach to identify the degree to which infrastructure assets are at risk and provide stakeholders and decision-makers with the suggestions on minimizing the effects of future flood events.
Get full access to this article
View all available purchase options and get full access to this chapter.
REFERENCES
ASCE. 2018. “2018 Michigan Infrastructure Report Card.” Accessed August 20, 2023. https://infrastructurereportcard.org/state-item/michigan/.
Buckman, S., Arquero de Alarcon, M., and Maigret, J. 2019. “Tracing shoreline flooding: Using visualization approaches to inform resilience planning for small Great Lakes communities.” Applied Geography, 113, 102097.
Dawson, R. J., Speight, L., Hall, J. W., Djordjevic, S., Savic, D., and Leandro, J. 2008. “Attribution of flood risk in urban areas.” Journal of Hydroinformatics, 10(4), 275–288.
Donnelly, J., Abolfathi, S., Pearson, J., Chatrabgoun, O., and Daneshkhah, A. 2022. “Gaussian process emulation of spatio-temporal outputs of a 2D inland flood model.” Water Research, 225, 119100.
Gallagher, G. E., Duncombe, R. K., and Steeves, T. M. 2020. “Establishing Climate Change Resilience in the Great Lakes in Response to Flooding.” Journal of Science Policy & Governance, 17(01).
Kundzewicz, Z. W., et al. 2014. “Flood risk and climate change: Global and regional perspectives.” Hydrological Sciences Journal, 59(1), 1–28.
López-Lopera, A. F., Idier, D., Rohmer, J., and Bachoc, F. 2022. “Multioutput Gaussian Processes with Functional Data: A Study on Coastal Flood Hazard Assessment.” Reliability Engineering & System Safety, 218, 108139.
NOAA (National Oceanic and Atmospheric Administration) National Centers for Environmental Information. 2023. Billion-Dollar Weather and Climate Disasters. Accessed on August 20, 2023. https://www.ncei.noaa.gov/access/billions/summary-stats/US/2020-2022.
Pickard, A., and Barlock, R. 2022. Partnering for Resilience: The Southeast Michigan Flooding Risk Tool. United States Department of Transportation. Accessed on August 20, 2023. https://highways.dot.gov/public-roads/spring-2022/02.
Rasmussen, C. E., and Williams, C. K. I. 2006. Gaussian processes for machine learning. (Vol. 1, p. 159). Cambridge, MA: MIT press.
Senouci, A., Elabbasy, M., Elwakil, E., Abdrabou, B., and Zayed, T. 2014. “A model for predicting failure of oil pipelines.” Structure and Infrastructure Engineering, 10(3), 375–387.
Information & Authors
Information
Published In
History
Published online: Mar 18, 2024
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.