W. Bangerth, M. Adams, N. Amato, S. Chirayath, J-L Guermond, G. Kanschat, P. Kuchment, J. Morel, J. Ragusa, L. Rauchwerger “Inverse Modeling to Detect Illicitly Smuggled Materials in Containers”, Poster presented at the Academic Research Initiative Grantees Conference, Washington, D.C., April 2008.
To improve the chances of detecting nuclear material, for example HEU, smuggled in cargo containers or cars, all available data needs to be used simultaneously, not only by specifying a threshold on each individual detector reading. We present a framework in which we can compute the probability for false positives and false negatives. Such a framework also allows us to derive optimization problems that determine operating parameters that minimize the probability of false negatives subject to constraints such as cost, delay, etc. Furthermore, it allows us to evaluate the impact of proposed new detector concepts.