Statistical learning techniques for the estimation of lifeline network performance and retrofit selection

Jason Wu and Jack W. Baker

Reliability Engineering and System Safety, 2020, vol. 200, issue C

Abstract: The reliability of water supply networks subjected to catastrophic events is a crucial concern to communities, but our ability to assess these systems is limited by their size and complexity. This paper proposes a statistical learning technique, Random Forests, to efficiently estimate network performance in place of direct physical simulation. This technique uses a set of explanatory metrics that describe the impact of seismic damage to network behavior. The approach is applied to a case study network, the Auxiliary Water Supply System of San Francisco. The resulting statistical model is shown to replicate network performance estimates from flow-based hydraulic simulation, and exhibits good performance in identifying components to retrofit to improve the reliability of the system. The favorable performance and computational advantages of this approach make it an attractive tool for infrastructure reliability and risk mitigation analyses.

Keywords: Statistical learning; Infrastructure reliability; Seismic risk; Seismic retrofitting (search for similar items in EconPapers)
Date: 2020
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0951832019306933
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:reensy:v:200:y:2020:i:c:s0951832019306933

DOI: 10.1016/j.ress.2020.106921

Access Statistics for this article

Reliability Engineering and System Safety is currently edited by Carlos Guedes Soares

More articles in Reliability Engineering and System Safety from Elsevier
Bibliographic data for series maintained by Catherine Liu ().