Abstract

We  describe  an  innovative  application  of  a  novel game-theoretic approach for a national scale secu-rity deployment.   Working with the United States Transportation Security Administration (TSA), we have developed a new application called GUARDS to allocate the TSA’s limited resources across hun-dreds  of  security  activities  to  provide  protection at  over  400  United  States  airports.   Similar  secu-rity applications (e.g., ARMOR and IRIS) have fo-cused on one-off tailored applications and one se-curity  activity  (e.g.   checkpoints)  per  application,GUARDS on the other hand faces three new key issues:  (i) reasoning about hundreds of heteroge-neous security activities; (ii) reasoning over diverse potential threats; (iii) developing a system designed for hundreds of end-users. Since a national deploy-ment  precludes  tailoring  to  specific  airports,  our key  ideas  are:   (i)  creating  a  new  game-theoretic framework that allows for heterogeneous defender activities and compact modeling of a large number of threats; (ii) developing an efficient solution tech-nique based on general purpose Stackelberg game solvers; (iii) taking a partially centralized approach for knowledge acquisition.   The scheduling assis-tant has been delivered to the TSA and is currently undergoing  evaluation  for  scheduling  practices  at an undisclosed airport.  If successful, the TSA in-tends  to  incorporate  the  system  into  their  unpre-dictable scheduling practices nationwide.