Super Cat Characterization Through Deconstruction of Stochastic Event Cascades

Abstract eng:
Losses generated by major catastrophes are generally estimated from a characterization of the damages caused by the (primary) event (an earthquake) without fully modeling the following secondary events (FFEQ, landslide, liquefaction, etc.), and some subjective amount of losses, with large uncertainty, is added to account for the un-modeled part. We have developed a method that models our sphere of living, working, and sustenance as a complex system. Some usually well-characterized elements (ie. Bridges, hospitals, interstate highways, emergency organization) constitute the well-defined network of nodes and links. The myriad of other small components, which have a small relative effect on the whole, but jointly can have a large importance, are modeled statistically. This method specifically accounts for the situations where more and more small secondary cascading events occur. Of great interest are the cases when the cascade of events causes a gradual degradation of the entire socio-economic matrix of a region, leading to a situation akin to a phase change in the behavior of the system. Such conditions generally lead to extreme losses, and are often referred to as Super-Cat. The progression of events in the hurricane Katrina of 2003 is one example of such a condition, in which the entire infrastructure and administrative matrix collapsed. Our method uses a deconstruction of historical catastrophic events, and we formulate a model of stochastic occurrence of cascades. We use the concept of disruption to quantify the effect of un-modeled events on the final losses.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket