Optimal Layered Defense For Site Protection

TL;DR

This paper develops a dynamic programming approach for optimizing layered security in site protection, addressing both standard and adaptive adversaries, with computational experiments demonstrating effectiveness.

Contribution

It introduces a novel dynamic programming method for layered security optimization and extends it to adaptive adversaries, filling a gap in site protection modeling.

Findings

Effective optimization of layered security configurations

Successful modeling of adaptive adversary behavior

Computational experiments validate the proposed methods

Abstract

We present a model for layered security with applications to the protection of sites such as stadiums or large gathering places. We formulate the problem as one of maximizing the capture of illegal contraband. The objective function is indefinite and only limited information can be gained when the problem is solved by standard convex optimization methods. In order to solve the model, we develop a dynamic programming approach, and study its convergence properties. Additionally, we formulate a version of the problem aimed at addressing intelligent adversaries who can adjust their direction of attack as they observe changes in the site security. Furthermore, we also develop a method for the solution of the latter model. Finally, we perform computational experiments to demonstrate the use of our methods.