Logic Trees: Fault, Success, Attack, Event, Probability, and Decision Trees

Abstract

This article describes the fundamentals of logic trees used in decision and risk analysis. We include in the broad definition of logic trees any conceptual or mathematical model that describes scenarios of events and/or decisions in a treelike structure. Logic trees are important tools for analyzing uncertain events, defining accident scenarios, assessing risk, calculating joint probability distributions, and evaluating decision options. Six types of logic trees are described, compared, and illustrated in this article: fault trees, success trees, attack trees, event trees, probability trees, and decision trees. Fault trees deductively model the combination of system failures and human errors that could lead to an accident. Success trees are the complement of fault trees and model the combination of events that lead to success. Attack trees model the sequential actions of an intelligent adversary to defeat a defensive system. Event trees inductively model the sequences of events that lead to an accident. Probability trees model a sequence of uncertain events in order to calculate the probability of events in the outcome space. Decision trees are logic trees that include sequential decisions in addition to sequential uncertain events. Examples are provided in this article to illustrate the similarities and differences among the types of logic trees. We conclude with challenges for applying logic trees to homeland security applications.

Using Incident Response Trees as a Tool for Risk Management of Online Financial Services

“…Compared to an IRT, the attack tree is a complementary model in the sense that it can be used to estimate the probability of top events. In a comprehensive risk analysis, the probability of the top event can then be connected to the initiating event in the IRT, (21) i.e., the attack tree analysis may provide some of the statistics needed for an IRT analysis. Compared to an IRT, the protection tree is a complementary model in that it can be used to estimate how countermeasures can be made more effective.…”

A review: Monitoring situational awareness of smart grid cyber‐physical systems and critical asset identification

“…Fault tree analysis (FTA) [122,123] � It focusses on a system's safety and reliability. � It can be used for observing the impact and likelihood of equipment failure.…”

Real‐time threat assessment based on hidden Markov models

“…Regarding the crucial issue of terrorism risk, there is no single definition that is widely adopted; in general, the notion of terrorism risk is related to threats, attacks, vulnerabilities, consequences of the attacks, uncertainties, and probabilities (Aven & Guikema, 2015). Given the above‐mentioned facets of terrorism risk, a discussion about the applicability of probabilistic approaches for threat assessment in such situations (Ezell et al., 2010) has included Logic Trees (e.g., Dillon‐Merrill et al., 2008) and Bayesian Network analysis (e.g., Hudson et al., 2005). Regression analysis has also been employed to model regional terrorism risk (Chatterjee & Abkowitz, 2011) taking into consideration the population density and the number of critical infrastructures in each region.…”