A Data-Driven Model to Generate Disruptive Scenarios for Infrastructure Resilience Studies

Devendra P. Jaiswal, Harsh Anand, Satish Mahade Van Srinivasan, Mohamad Darayi

Research output: Contribution to journalConference articlepeer-review

3 Scopus citations


This work proposes a data-driven model that uses data on natural disasters from the National Oceanic and Atmospheric Administration (NOAA) to enable predictive analytics and simulation of disruptions caused by natural hazards. Random generation of disruption is something that would be simple to implement. Still, it might not help researchers/policy-makers test the resilience from the occurrence of a natural disaster standpoint. Natural hazards such as a hurricane, tornado, or tropical storm are usually uncertain to predict unless we deploy a complex prediction algorithm that takes different atmospheric variables into account. Rather than diving into the meteorological predictive techniques, a data-driven model is proposed using the apriori algorithm to rely on historical data. Historical data-based techniques help generate disruption scenarios based on their historical occurrence and their topographical propagation. This, in turn, gives us disruptions that are historically significant and possible from their occurrence perspective.

Original languageEnglish (US)
Pages (from-to)248-255
Number of pages8
JournalProcedia Computer Science
StatePublished - 2021
Event2021 Complex Adaptive Systems Conference - Malvern, United States
Duration: Jun 16 2021Jun 18 2021

All Science Journal Classification (ASJC) codes

  • General Computer Science


Dive into the research topics of 'A Data-Driven Model to Generate Disruptive Scenarios for Infrastructure Resilience Studies'. Together they form a unique fingerprint.

Cite this