Answers are grounded in this paper and cite their sources.
Quantitative data relating to interdependencies in many infrastructure networks are inadequate or not easily available due to several reasons, such as security, business concerns, and a lower level of technology adoption. Alternatively, past studies recommend eliciting imprecise information about infrastructure dependencies from experts. Though collecting subjective information is comparatively easier, their use in quantitative models gives rise to epistemic uncertainty, which could significantly impact resilience investment decisions. This study investigates the use of imprecise dependency information, or more specifically linguistic descriptions of dependencies, to develop a hybrid metric to evaluate the event-specific network-wide vulnerabilities of infrastructure failures. In order to handle the epistemic uncertainties associated with the linguistic dependencies, the network-wide impacts of extreme events is represented using a pair of extreme probability distributions (derived using possibility theory) and a most-likely distribution. The study also demonstrates how reducing epistemic uncertainties is key in interdependent infrastructure models to obtain actionable results.
No takes yet. Share an insight, caveat, or question.
Balakrishnan et al. (2020) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: