Quantitative risk-based analysis for military counterterrorism systems

This paper presents a realistic and practical approach to quantitatively assess the risk-reduction capabilities of military counterterrorism systems in terms of damage cost and casualty figures. The comparison of alternatives is thereby based on absolute quantities rather than an aggregated utility or value provided by multicriteria decision analysis methods. The key elements of the approach are (1) the use of decision-attack event trees for modeling and analyzing scenarios, (2) a portfolio model approach for analyzing multiple threats, and (3) the quantitative probabilistic risk assessment matrix for communicating the results. Decision-attack event trees are especially appropriate for modeling and analyzing terrorist attacks where the sequence of events and outcomes are time-sensitive. The actions of the attackers and the defenders are modeled as decisions and the outcomes are modeled as probabilistic events. The quantitative probabilistic risk assessment matrix provides information about the range of the possible outcomes while retaining the simplicity of the classic safety risk assessment matrix based on Mil-Std-882D. It therefore provides a simple and reliable tool for comparing alternatives on the basis of risk including confidence levels rather than single point estimates. This additional valuable information requires minimal additional effort. The proposed approach is illustrated using a simplified but realistic model of a destroyer operating in inland restricted waters. The complex problem of choosing a robust counterterrorism protection system against multiple terrorist threats is analyzed by introducing a surrogate multi-threat portfolio. The associated risk profile provides a practical approach for assessing the robustness of different counterterrorism systems against plausible terrorist threats. The paper Regular Paper *This is a US Government work, and, as such, is in the public domain in the United States of America. †Author to whom all correspondence should be addressed (e-mail: [email protected]). Systems Engineering, Vol. 10, No. 4, 2007 © 2007 Wiley Periodicals, Inc.