Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Alliance Grants

McGill University

$25,000.00

Physics-Informed Machine Learning for Regional Resilience Quantification

FY 2023-24

Summary
Resilience is defined as the capacity of infrastructure, communities, and their related systems to mitigate, adapt, or positively respond to chronic and acute stresses (e.g., natural hazards such as earthquakes), transforming in ways that restore, maintain, and even improve their essential functions. The objective of the proposed international collaboration is to develop the Physics-Informed Machine Learning Model for Regional Resilience Quantification (PIML4ResQ), an artificial intelligence (AI)-aided decision support tool designed to quantify and enhance the resilience of Canada's infrastructure. Assessing the resilience of regional infrastructure via simulations demands significant computational resources, encompassing various elements like hazard analysis, exposure modeling, fragility assessment, and consequence evaluation. On the other hand, quantifying resilience using machine learning (ML) relies heavily on vast datasets for model training. Therefore, it is crucial to effectively merge these approaches through PIML, which achieves a balanced blend of computational efficiency and model accuracy. Also, purely data- or simulation-driven resilience models lack the needed capacity for generalization and consistency, while PIML stays at the nexus that can move from diagnosis to prognosis. The PIML4ResQ proposed from the current project will be tailored for eastern Canada to respond to earthquake risks with unmatched effectiveness (e.g., resource allocation, emergency response, retrofit prioritization); it will also position Canada a global leader to assess and enhance community resilience.
Program
Alliance Grants
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
591236-2023
Name nom
Xie, Yazhou Y
Department département
Civil Engineering and Applied Mechanics
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
25000
Program id
ALLRP
Program name en
Alliance Grants
Program name fr
Subventions Alliance
Group en
RESEARCH PARTNERSHIPS
Group fr
PARTENARIATS DE RECHERCHE
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
601
Area of application group en
Construction, urban and rural planning
Area of application group fr
Construction, urbanisme et aménagement rural
Area of application en
Structural engineering
Area of application fr
Génie des charpentes
Research subject code
1107
Research subject group en
Structural engineering
Research subject group fr
Earthquake engineering
Research subject en
Génie des structures
Research subject fr
Génie parasismique
Application title
Physics-Informed Machine Learning for Regional Resilience Quantification
Keyword
Bridge Network; Infrastructure Resilience; Machine Learning; Physics-based Constraints; Seismic Resilience
Application summary
Resilience is defined as the capacity of infrastructure, communities, and their related systems to mitigate, adapt, or positively respond to chronic and acute stresses (e.g., natural hazards such as earthquakes), transforming in ways that restore, maintain, and even improve their essential functions. The objective of the proposed international collaboration is to develop the Physics-Informed Machine Learning Model for Regional Resilience Quantification (PIML4ResQ), an artificial intelligence (AI)-aided decision support tool designed to quantify and enhance the resilience of Canada's infrastructure. Assessing the resilience of regional infrastructure via simulations demands significant computational resources, encompassing various elements like hazard analysis, exposure modeling, fragility assessment, and consequence evaluation. On the other hand, quantifying resilience using machine learning (ML) relies heavily on vast datasets for model training. Therefore, it is crucial to effectively merge these approaches through PIML, which achieves a balanced blend of computational efficiency and model accuracy. Also, purely data- or simulation-driven resilience models lack the needed capacity for generalization and consistency, while PIML stays at the nexus that can move from diagnosis to prognosis. The PIML4ResQ proposed from the current project will be tailored for eastern Canada to respond to earthquake risks with unmatched effectiveness (e.g., resource allocation, emergency response, retrofit prioritization); it will also position Canada a global leader to assess and enhance community resilience.