Using Game Theory for Los Angeles Airport Security

n Security at major locations of economic or political impor tance is a key concern around the world, particularly given the threat of terrorism. Limited security resources prevent full security coverage at all times, which allows adversaries to observe and exploit patterns in selective patrolling or mon itoring; for example, they can plan an attack avoiding existing pa trols. Hence, randomized patrolling or monitoring is impor tant, but randomization must provide distinct weights to dif ferent actions based on their complex costs and benefits. To this end, this article describes a promising transition of the lat est in multiagent algorithms into a deployed application. In particular, it describes a software assistant agent called AR MOR (assistant for randomized monitoring over routes) that casts this patrolling and monitoring problem as a Bayesian Stackelberg game, allowing the agent to appropriately weigh the different actions in randomization, as well as uncertainty over adversary types. ARMOR combines two key features. It uses the fastest known solver for Bayesian Stackelberg games called DOBSS, where the dominant mixed strategies enable randomization; and its mixed-initiative-based interface allows users occasionally to adjust or override the automated schedule based on their local constraints. ARMOR has been successfully deployed since August 2007 at the Los Angeles International Airport (LAX) to randomize checkpoints on the roadways entering the airport and canine patrol routes within the airport terminals. This article examines the information, design choices, challenges, and evaluation that went into de signing ARMOR.