200 likes | 384 Views
Simulation and Analysis of Cascading Failure in Critical Infrastructure. Robert Glass 1 , Walt Beyeler 1 , Kimmo Soram ä ki 2 , Morten Bech 3 and Jeffrey Arnold 3 1 Sandia National Laboratories 2 European Central Bank 3 Federal Reserve Bank of New York. “heavy tail” region.
E N D
Simulation and Analysis of Cascading Failure in Critical Infrastructure Robert Glass1, Walt Beyeler1, Kimmo Soramäki2, Morten Bech3 and Jeffrey Arnold3 1Sandia National Laboratories 2European Central Bank 3Federal Reserve Bank of New York
“heavy tail” region First Stylized Fact: Multi-component Systems often have power-laws & “heavy tails” “Big” events are not rare in many such systems Earthquakes: Guthenburg-Richter Extinctions, Forest fires log(Frequency) Wars, Epidemics, Cities Traffic jams, Stock crashes Power Blackouts Telecom outages log(Size)
Dissipation Power Law - Critical behavior – Phase transitions Equilibrium systems: e.g., Magnets & the Curie point What keeps a non-equilibrium system at a phase boundary? Correlation External Drive Temperature Tc
Drive Relaxation Cascade from Local Rules 1987 Bak, Tang, Wiesenfeld’s “Sand-pile” or “Cascade” Model Lattice “Self-Organized Criticality” power-laws fractals in space and time
Second Stylized Fact: Networks are Ubiquitous Food Web Molecular Interaction New York state’s Power Grid Illustrations of natural and constructed network systems from Strogatz [2001].
Special properties of the “Scale-free” network Power-law degree distribution Hierarchical with “King-pin” nodes Properties: vulnerable to informed attack… tolerant to random failure
Other Networks Nodes Links Tailored Interaction Rules Actors Drive Dissipation Our Conceptual Approach: Rules ON Networks for Bottoms up Simulation of Infrastructures PolyNet Built in Repast Network Adapt & Rewire
Development & Applications Abstract Studies Stylized Physical Infrastructure Applications: High Voltage Electric Power Grids Payment and Banking Systems Information Networks Physical + SCADA + Market + Policy Forcing Stylized Social Applications: Epidemics Social/Report Network Evolution Self-organized Terrorist/Extremist groups Crisis and recovery from WMD & Bio attacks Where we are headed: Combined Physical-Human “Infrastructure” Systems
Fish-net or Donut size time Scale-free log(freq) log(size) BTW sand-pile on varied topology Random sinks Sand-pile rules and drive 10,000 nodes
Fish-net Fish-net or Donut size time Scale-free Scale-free size time Cascading Blackouts Sources, sinks, relay stations, 400 nodes DC circuit analogy, load, safety factors Random transactions between sources and sinks
Balance 0 Opening balance adapts to control risk 0 Time Trading Day Pay Balance Training Period Cascading Period Balance 0 Time Cascading Liquidity Loss within Payment Systems banks payments
Patterned Transactions Random removal vs Attack of the Highest Degree node bank defaults time liquidity time Cascading Liquidity in Scale-free Network
Agent classes Teen Laura Glass’s Groups me Links & Frequency me Extended Family me Teen Extra Class Specific Parameters me me • Infectivity • Mortality • Immunity • Etc. Nuclear Family Classes (There are 6 of these) Kids Teens Everyone Adults Seniors Cascading Infectious Diseases Parameters can change when Symptomatic 13
Influenza Epidemic in Structured Village of 10,000: Increasing Realism Without Immunity Agent differentiation With Immunity & Mortality Behavioral Changes when Symptomatic Structure: Heterogeneous Network + Like with Like 14
Seniors only (yellow) Current policy Kids & Teens! Flu Epidemic Mitigation: Vaccination Strategies <60% required Network Structure + Physics of Transmission Process Allows Effective Mitigation Design
General Remarks: Concepts from Complexity Science are valuable and allow a simulation approach for critical infrastructures that is flexible and has wide ranging applications Focus on POLICY Detailed applications with Domain experts Developmental directions Generalization/Abstraction Encapsulation/Integration -NABLE -BOF simulator Tools/Insight for Rapid deployment