Natural hazards have the potential to cause large scale impacts and disruption to all countries and if these events occur in highly populated areas the impacts can be catastrophic. The severity and lasting impact of these hazards are often linked to the resilience of critical infrastructure systems (including: water distribution networks, electrical systems and transportation networks) which underpin our communities and support social and economic development.
The seminar I presented on the 21st August at SMART Infrastructure Facility reported aspects of my research using network theory and catastrophe risk modelling techniques to assess the resilience of real-world infrastructure systems. Traditionally, the resilience of infrastructure systems is assessed by subjecting physically based models to a range of hazard scenarios. However, this approach can only inform us of inadequacies in the system for the chosen scenarios, potentially leaving us vulnerable to unforeseen events. In this seminar, I presented my research using network graph theory to assess the resilience of infrastructure systems. This analysis technique has the ability to give a level of confidence that the system will perform well to unforeseen events. I also showed how network theory metrics and traditional physically based measures can be combined to identify specific “vulnerable” components within an infrastructure system; (these “vulnerable” components being those that have a disproportionate impact to the functioning of the remaining network if they fail).
I also presented recent research work using a catastrophe risk modelling framework to assess the impact of windstorm hazards to electrical distribution infrastructure in the UK. In simple terms, a catastrophe model assesses risk by combining an exposure database with a hazard model, a vulnerability model, and an economic loss model. The initial results from this approach, show that it can be used to assess the resilience of infrastructure systems, provided that a system performance model is combined in the analysis.