Abstract
The problem of container inspection at ports-of-entry is formulated in several different ways as an optimization problem. Data generated from different analytical methods, x-ray detectors, gamma-ray detectors and other sensors used for the detection of chemical, biological, radiological, nuclear, explosive, and other illicit agents are often relied upon to make critical decisions with regard to the nature of containers presented for possible inspection and the appropriate response mechanism. Several important questions related to the utilization and coordination of multiple sensors for container inspection are discussed. New and efficient algorithms for finding the best inspection strategy, including the optimal sequencing of sensors and optimal assignment of thresholds for interpreting sensor readings, are described. Models and algorithms that can be used by decision makers, allowing them to minimize expected cost of inspection, minimize inspection errors (both false positives and false negatives), and/or maximize the throughput of containers, are outlined.