Evaluating Risk from Acts of Terrorism with Belief and Fuzzy Sets

Abstract

Risk consists of the likelihood of an event combined with the consequence of that event. There is uncertainty associated with an estimate of risk for an event that may happen in the future. For random, "dumb" events, such as an earthquake, this uncertainty is aleatory (stochastic) in nature and can be addressed with the probability measure of uncertainty. A terrorist act is not a random event; it is an intentional act by a thinking malevolent adversary. Much of the uncertainty in estimating the risk of a terrorist act is epistemic (state of knowledge); the adversary knows what acts will be attempted, but we as a defender have incomplete knowledge to know those acts with certainty. To capture the epistemic uncertainty in evaluating the risk from acts of terrorism, we have applied the belief/plausibility measure of uncertainty from the Dempster/Shafer theory of evidence. Also, to address how we as a defender evaluate the selection of scenarios by an adversary, we have applied approximate reasoning with fuzzy sets. We have developed software to perform these evaluations