Large-scale natural and human-initiated crises that affect significant population centers are important from the standpoint of national security, resilient design of cities, and the overall security and well-being of society.

Summary

The focus of this work is on human-initiated crises, although many of our broad conclusions are likely to be applicable in more general settings. Unlike natural disasters, our understanding of the effects of large-scale human initiated crises is limited; we are fortunate not to have experienced too many such events as global citizens. Nevertheless, planning for such scenarios is vitally important in order to be prepared should such an event actually occur. Computer simulations and an associated modeling and informatics environment provide a natural analytical tool that can aid in developing and assessing various policies and response plans in the event of such crises. Here we focus on a hypothetical scenario published by the Department of Homeland Security, which is essentially similar to a scenario that has been studied for the past 60 years. In this scenario, an improvised nuclear device (IND) is detonated in a large urban region.

The Scenario

Unannounced 10 kt detonationBuilding damage and fallout path

  • Location:
    • -77.036574 longitude
    • -38.902604 latitude
  • Washington DC
  • Time: 11:15 EDT
  • Date: May 15, 2006

Methods

We built detailed individual-level models of the population of the region, human behaviors in the aftermath of disasters, and four major infrastructures: transportation, communication, health, and power.

Interactions between human behavior and infrastructuresHuman behavior interacts with infrastructures in complex feedback loops. Individual-level modeling allows us to incorporate models of human behavior derived from the literature and from surveys and to understand how individual behaviors and actions lead to emergent large-scale effects on the infrastructure. For example, mass evacuation can lead to traffic jams, especially if roads are damaged, which can lead to communication failure as cellphone base stations get overloaded. Lack of communication, in turns, leads to changes in behavior.

These models are built using synthetic information technology that has been pioneered and developed over more than 20 years in the Network Dynamics and Simulation Science Laboratory. The figures below show multiple sources of information are integrated and synthesized for the simulation and how information is updated as the simulation proceeds.

 

Synthetic data generationSystem data flow

 

 

 

 

 

 

Using high-performance computing resources and a novel database-driven simulation architecture we carried out a number of simulations of multiple scenarios and interventions.

Lessons Learned

  • The first 3 day problem is different than the 3 weeks, 3 months or 3 years problem
    • Behaviors, response & state assessment problem in first 3 days is different than the problem in later stages
    • E.g., information-seeking and shelter-seeking behaviors that are important to represent in the first 3 days are not really as relevant later. Similarly housing and economic problems that are critical later are not important early on
  • Seemingly minor mitigation effects (as compared to the event itself) have effects that are observable in behavioral details.
    • Will play an important role in setting up the longer term problem
    • Communication and information dissemination is central to state assessment, coordination, and to individual survival. It can reduce panic, improve compliance to sheltering, support household reconstitution
  • Has to be understandable, efficient to compute & capture infrastructure effects
    • "Behavioral Backbone"
    • Individual and collective behavior and behavioral adaptation
    • Sensitive to contextual information
  • Detailed behavioral representation and its co-evolution with the physical infrastructure has subtle but important effects and implications
    • Produce aggregate statistics that are consistent with previous studies, BUT
    • They also
      • provide important details on spatial distribution of populations
      • allow us to represent and track important sub-populations, e.g. elected federal officials, critical workers, etc.
      • allow us to understand the contextual information and yet lead to formulation of general guidelines
  • Importance of modeling individuals, groups and critical populations
    • Migration pathways are complex emergent social behaviors in socio-technical catastrophe contexts
    • Responder behavior affected by evacuation routes
    • Government functioning and recovery affected (need to represent government officials)
    • Studies that ignore behaviors like household reconstitution will generate overly optimistic estimates of evacuation times and radiation exposure
  • Lead to formulation of conclusions that are likely to be applicable in more general contexts
    • E.g. rapid deployment of communication network is an important response strategy: will likely save lives, reduce adverse health effects and lead to reduction in worry.

Publications

Bryan L. Lewis, Samarth Swarup, Keith R. Bisset, Stephen G. Eubank, Madhav V. Marathe, Christopher L. Barrett (2013), A Simulation Environment for the Dynamic Evaluation of Disaster Preparedness Policies and Interventions, Journal of Public Health Management and Practice, 19(Suppl 5): S42-8.

Madhav V. Marathe, Henning S. Mortveit, Nidhi Parikh, and Samarth Swarup (2014), Prescriptive Analytics Using Synthetic Information, In, William H. Hsu (Ed.), Emerging Trends in Predictive Analytics: Risk Management and Decision Making, IGI Global: Hershey, PA.

Christopher L. Barrett, Virgilio Centeno, Stephen G. Eubank, C. Yaman Evrenosoglu, Achla Marathe, Madhav V. Marathe, Chetan Mishra, Henning S. Mortveit, Anamitra Pal, Arun Phadke, Jim Thorp, Anil Vullikanti and Mina Youssef (2014), Impact of Surface Nuclear Blast on the Transient Stability of the Power System, The 9th International Conference on Critical Information Infrastructures Security, Limassol, Cyprus, October 13-15.

Christopher L. Barrett, Keith R. Bisset, Shridhar Chandan, Jiangzhuo Chen, Youngyun Chungbaek, Stephen G. Eubank, C. Yaman Evrenosoglu, Bryan L. Lewis, Kristian Lum, Achla Marathe, Madhav V. Marathe, Henning S. Mortveit, Nidhi Parikh, Arun Phadke, Jeffery Reed, Caitlin Rivers, Sudip Saha, Paula Stretz, Samarth Swarup, James Thorp, Anil Vullikanti, Dawen Xie (2013), Planning and Response in the Aftermath of a Large Crisis: An Agent-based Informatics Framework, The Winter Simulation Conference, Washington DC, USA, Dec 8-11.

Samarth Swarup, Kristian Lum, Christopher L. Barrett, Keith R. Bisset, Stephen G. Eubank, Madhav V. Marathe, Paula Stretz (2013), A Synthetic Information Approach to Urban-scale Disaster Modeling, The 2nd International Conference on Big Data Science and Engineering (BDSE 2013), Sydney, Australia, Dec 3-5.

Christopher L. Barrett, Stephen G. Eubank, C. Yaman Evrenosoglu, Achla Marathe, Madhav V. Marathe, Arun Phadke, James Thorp, Anil Vullikanti (2013), Effects of Hypothetical Improvised Nuclear Detonation on the Electrical Infrastructure, Invited paper at the CRIS meeting at Internationaler ETG-Kongress, Berlin.

Abhijin Adiga, Madhav V. Marathe, Henning S. Mortveit, Sichao Wu, Samarth Swarup (2013), Modeling Urban Transportation in the Aftermath of a Nuclear Disaster: The Role of Human Behavioral Responses, The Conference on Agent-Based Modeling in Transportation Planning and Operations, Blacksburg, VA, Sep 30 - Oct 2.

Nidhi Parikh, Samarth Swarup, Paula Stretz, Caitlin Rivers, Bryan L. Lewis, Madhav V. Marathe, Stephen G. Eubank, Christopher L. Barrett, Kristian Lum, Youngyun Chungbaek (2013), Modeling Human Behavior in the Aftermath of a Hypothetical Improvised Nuclear Detonation, In Proceedings of The Twelfth International Conference on Autonomous Agents and Multiagent Systems (AAMAS), St. Paul, MN, USA, May, pp. 949-956.

Abhijin Adiga, Henning S Mortveit, Sichao Wu (2013), Route Stability in Large-scale Transportation Models, The Workshop on Multiagent Interaction Networks (MAIN), held in conjunction with the 12th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), St. Paul, MN, USA, May 7.

Shridhar Chandan, Sudip Saha, Christopher L. Barrett, Stephen G. Eubank, Achla Marathe, Madhav V. Marathe, Samarth Swarup and Anil Kumar Vullikanti (2013), Modeling the Interactions between Emergency Communications and Behavior in the Aftermath of a Disaster, The International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction (SBP), Washington DC, USA, April 2-5.

Talks

Samarth Swarup (2013), Prescriptive Analytics for Urban Resilience, Oral presentation at the AAAI Fall Symposium on Social Networks and Social Contagion, Nov 15.
 
Christopher L. Barrett, Stephen G. Eubank, C. Yaman Evrenosoglu, Achla Marathe, Madhav V. Marathe, Arun Phadke, Jim Thorp and Anil Vullikanti (2013), Effects of Hypothetical Improvised Nuclear Detonation on the Electrical Infrastructure, Keynote talk at the CRIS meeting at ETG Berlin, November.
 
Christopher L. Barrett (2013), Policy Informatics at Societal Scale: Massively Interactive Socially-Coupled Systems, 4th Conference on Community Resiliency, Davos, Switzerland, August 29-30.
 
Christopher L. Barrett (2013), Policy Informatics at Societal Scale: Massively Interactive Socially-Coupled Systems, at the session on Global City Systems: How a "Global Systems Science" Can Help Cope with Urbanisation, European Forum Alpbach, Austria, August 26.
 
Kristian Lum (2013), Bayesian Spatial Quantile Regression Using the Asymmetric Laplace Process, Invited talk at The Joint Statistical Meetings (JSM). Montreal, Canada, August 5.

Samarth Swarup (2012), Simulating Human Behavior in the Aftermath of a Catastrophe, Oral presentation at the AAAI Fall Symposium on Social Networks and Social Contagion, Nov 2.

Caitlin Rivers (2012), Behavior-driven Simulations of Disasters to Foster Resilient Communities, Oral presentation at the 140th Annual Conference of the American Public Health Association, Oct 27-30, San Francisco, CA.

Bryan L. Lewis (2012), Dynamic Policy Evaluation Environment For Disaster Preparedness, Oral presentation at the Dynamics of Preparedness Conference, Oct 22-24, Pittsburgh, PA.

Christopher L. Barrett and Madhav V. Marathe (2012), Planning and Responding to Human Initiated Crisis: Role of Data Intensive Computing and Computational Socio-Technical Sciences, International Conference on Networks in Biology, Social Science and Engineering, Bangalore, India, July.
 
Madhav Marathe (2012), Social, Health and Socio-technical effects of an IND in the National Capitol: Advances in Data Intensive Computing and Simulation Sciences, MIDAS meeting, June.

Back to top