Cause-and-Effect or Fooled by Randomness?
Loading...
Files
Authors
Lewis, Ted G.
Subjects
Advisors
Date of Issue
2010-01-00
Date
2010-01
Publisher
Monterey, California. Naval Postgraduate School
Center for Homeland Defense and Security
Center for Homeland Defense and Security
Language
en_US
Abstract
This article develops an unconventional theory of infrastructure criticality based on decade-old ideas from a variety of disciplines. First, the concept of self-organized criticality (SOC) is explained using three simple simulations proposed by Per Bak, Newman, and Amaral-Meyers. Each simulation illustrates an aspect of SOC: self-organization, randomness as an underlying engine of disaster, and the role of interdependency or connectivity in complex systems. Next, the discussion shifts to an explanation of a general property shared by many major disasters: the fractal power law. Power laws turn out to be appropriate proxies for the insurance industry measure of likelihood called exceedence probability (probability of consequence equal to or greater than some size). The power law exceedence probability curve is associated with nearly all sectors prone to catastrophe. This is no coincidence, but more intriguing is the realization that power law exceedence probability curves can be produced from purely underlying randomness. This supports the author's conjecture that catastrophic incidents (often) occur because of randomness -- not strictly cause-and-effect.
Type
Article
Description
This article appeared in Homeland Security Affairs (January 2010), v.6 no.1
Series/Report No
Department
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
Citation
Homeland Security Affairs (January 2010), v.6 no.1
Distribution Statement
Rights
The copyright of all articles published in Homeland Security Affairs rests with the author[s] of the articles. Any commercial use of Homeland Security Affairs or the articles published herein is expressly prohibited without the written consent of the copyright holder. Anyone can copy, distribute, or reuse these articles as long as the author and original source are properly cited.