Risk-based grouping for checked baggage screening systems

With millions of bags checked at over 7000 baggage screening locations in US daily, the checked baggage screening system may be exploited by the terrorists to do harm to the homeland security. Due to such a huge amount of luggage, how to improve the effectiveness and efficiency of the screening system becomes a challenging task. In this paper, we propose a risk-based cost-effectiveness model where checked bags are classified into several risk classes according to their risk characteristics. According to their risk levels, bags from different classes may go through different device combinations sequentially. For a multiple-device screening system, we determine the optimal sequence of the screening devices and the separate grouping strategies for bags from different risk classes with the objective of minimizing the expected cost per bag. Based on a detailed numerical study, we compare our model with three other cost-effectiveness models (the first model assumes that there is only one risk class, the second model assumes that there is only one group for each risk class, and the third model assumes that all devices in a device combination need to be gone through). Our major conclusions are that our proposed model is beneficial compared with other three models and moreover, the relative benefit becomes larger when the authority commands a stricter upper bound for the probability of false clear.