Stop nuclear smuggling through efficient container inspection

• Main Authors: WANG, Xinrun, GUO, Qingyu, AN, Bo
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2017
• Subjects: Container inspection; Game theory; Nuclear smuggling; Artificial Intelligence and Robotics; Theory and Algorithms; Transportation
• Online Access: https://ink.library.smu.edu.sg/sis_research/9166; https://ink.library.smu.edu.sg/context/sis_research/article/10169/viewcontent/AAMAS17_Container_av.pdf
• Institution: Singapore Management University

Since 2003, the U.S. government has spent $850 million on the Megaport Initiative which aims at stopping the nuclear smuggling in international container shipping through advanced inspection facilities including Non-Intrusive Inspection (NII) and Mobile Radiation Detection and Identification System (MRDIS). Unfortunately, it remains a significant challenge to efficiently inspect more than 11.7 million containers imported to the U.S. due to the limited inspection resources. Moreover, existing work in container inspection neglects the sophisticated behavior of the smuggler who can surveil the inspector’s strategy and decide the optimal (sequential) smuggling plan. This paper is the first to tackle this challenging container inspection problem, where a novel Container Inspection Model (CIM) is proposed, which models the interaction between the inspector and the smuggler as a leader-follower Stackelberg game and formulates the smuggler’s sequential decision behavior as a Markov Decision Process (MDP). The special structure of the CIM results in a non-convex optimization problem, which cannot be addressed by existing approaches. We make several key contributions including: i) a linear relaxation approximation with guarantee of solution quality which reformulates the model as a bilinear optimization problem, ii) an algorithm inspired by the Multipleparametric Disaggregation Technique (MDT) to solve the reformulated bilinear optimization, and iii) a novel iterative algorithm to further improve the scalability. Extensive experimental evaluation shows that our approach can scale up to realistic-sized problems with robust enough solutions outperforming heuristic baselines significantly.