1 Optimal Allocation of Resources for Defense of Simple Series and Parallel Systems from Determined Adversaries Vicki M. Bier 1 and Vinod Abhichandani 2 Abstract Managing the risks posed by an intelligent and adaptable adversary is different from many other types of risk management. Thus, risk management in this context is a problem of game theory rather than decision theory. In other words, the defender wishes to choose the optimal strategy for defending against an optimal attack, and vice versa. In this paper, we apply game theory to help in characterizing optimal defensive strategies against intentional attacks. The results yield useful insights. Introduction In the aftermath of the September 11 th , 2001, attacks on the World Trade Center and the Pentagon (and the anthrax attacks in the United States), there is increased interest in strategies for protecting assets of value (including human life) against attacks by an intelligent and adaptable adversary. Even before that time, there were calls for greater attention to critical infrastructure protection, including computer security; see, e.g., President’s Commission on Critical Infrastructure Protection [1997]. This is a fundamentally different challenge from protecting against “acts of nature” or “accidents.” For example, an earthquake will not become stronger or “smarter” just because we have hardened our buildings to protect against it. In contrast, an intelligent and determined adversary is likely to adopt a different offensive strategy once we have put a particular set of protective measures in place. Therefore, good defensive strategies must consider the adversary’s behavior. 1 Professor, Dept. of Industrial Engineering, University of Wisconsin-Madison, 1513 University Avenue, Madison WI 53706; 608-262-2064; bier@engr.wisc.edu 2 Graduate student, Dept. of Industrial Engineering, University of Wisconsin-Madison, 1513 University Avenue, Madison, WI 53706; 608-263-2687; vinod_abhi@yahoo.com