Risk Analysis DOI: 10.1111/j.1539-6924.2011.01590.x A Comparative Analysis of PRA and Intelligent Adversary Methods for Counterterrorism Risk Management Jason Merrick 1, * and Gregory S. Parnell 2,3 In counterterrorism risk management decisions, the analyst can choose to represent terror- ist decisions as defender uncertainties or as attacker decisions. We perform a comparative analysis of probabilistic risk analysis (PRA) methods including event trees, influence dia- grams, Bayesian networks, decision trees, game theory, and combined methods on the same illustrative examples (container screening for radiological materials) to get insights into the significant differences in assumptions and results. A key tenent of PRA and decision analysis is the use of subjective probability to assess the likelihood of possible outcomes. For each technique, we compare the assumptions, probability assessment requirements, risk levels, and potential insights for risk managers. We find that assessing the distribution of potential attacker decisions is a complex judgment task, particularly considering the adaptation of the attacker to defender decisions. Intelligent adversary risk analysis and adversarial risk analysis are extensions of decision analysis and sequential game theory that help to decompose such judgments. These techniques explicitly show the adaptation of the attacker and the resulting shift in risk based on defender decisions. KEY WORDS: Adaptive adversary; risk management; terrorism risk 1. INTRODUCTION Decision trees, event trees, Bayesian networks, and influence diagrams have been widely applied to counterterrorism decision making. Von Winter- feldt and O’Sullivan, (1) Bakir, (2) and Merrick and McLay (3) apply such tree solutions to counterterror- ism. Game theory has also been applied in coun- terterrorism to model the decisions of the terrorist adversary, including general strategy and defense modeling (4−8) and specific applications, such as pro- tecting power transmission systems (9) and com- 1 Statistical Sciences and Operations Research, Virginia Common- wealth University, VA, USA. 2 Department of Systems Engineering, U.S. Military Academy at West Point, NY, USA. 3 Innovative Decisions Inc., Vienna, VA, USA. ∗ Address correspondence to Jason Merrick, PO Box 843083, 1015 Floyd Ave., Richmond, VA 23284-3083, USA; jmerric@vcu.edu. mercial airlines. (10) Pat ´ e-Cornell and Guikema (11) were the first to attempt to combine game theo- retic and decision analysis approaches fo counter- terrorism. More recently, Parnell et al. (12) and Rios Insua et al. (13) developed combinations of the two ap- proaches. Ezell et al. (14) reviewed a wide variety of models using different illustrative examples. In the following sections, we will review event trees, de- cision trees, Bayesian networks, influence diagrams, game theory, intelligent adversary risk analysis, and adversarial risk analysis and consider each applica- tion on the same two illustrative example decisions. We use the same examples and parameters for each model to allow comparison of the results under each approach and the decisions prescribed for the de- fender and the attacker. This approach enables a clearer comparison. Consider a simple counterterrorism decision where the defender has n alternatives, represented 1 0272-4332/11/0100-0001$22.00/1 C  2011 Society for Risk Analysis