Airline Security Through Artificial Intelligence
How the Transportation Security Administration Can Use Machine Learning to Improve the Electronic Baggage Screening Program
The Aviation and Transportation Security Act (Pub. L. 107-71, 2001) requires that the Transportation Security Administration (TSA) screen 100 percent of checked baggage at airports. To fulfill this mandate, TSA procures, installs, and maintains screening systems for checked baggage at airports through its Electronic Baggage Screening Program (EBSP). The Homeland Security Operational Analysis Center was asked to produce a 20-year EBSP technology roadmap to inform the future of checked baggage screening.
One possibility for improved baggage screening is to incorporate recent advances in artificial intelligence (AI) and machine learning (ML) in the screening process. There is a large and growing list of cognitive tasks for which ML can outperform humans, and many of these tasks are similar to those involved in TSA's screening process for checked baggage. The authors of this Perspective describe the current state, challenges, and outlook for AI and ML, using the EBSP example as a case study. The authors build on research and analysis conducted more broadly on technology development for EBSP, and much of this Perspective is drawn from the non–publicly available documentation of that work. The findings will likely be of interest to those planning to incorporate AI or ML into legacy technology systems.
Research conducted by
- Homeland Security Operational Analysis Center
HSOAC is a federally funded research and development center operated by the RAND Corporation under contract with DHS.