BAG-INTEL · Intelligent system for improved efficiency and effectiveness Intelligent system for improved efficiency and effectiveness of the customs control of passenger baggage from international flight arrivals

The BAG-INTEL project will provide robust AI based information utilization and decision support tools, within the context of advanced detection systems to support customs for increased effectiveness and efficiency of the customs control of air traveller baggage in inland border airports, while minimizing the human customs resources needed. This aim addresses the challenge of maintaining effective and efficient customs control of passenger baggage in the situation of the substantial growth of the volume of air travellers arriving in inland border airports with the limited human customs resources available. For this aim, the project will develop an integrated system solution comprising: (1) new AI powered functionality for enhanced detection of contraband in x-ray scanning of luggage, (2) AI camera based end-to-end reidentification of luggage, (3) digital twin for system visualisation and performance optimization for the operational context of an airport, (4) use case for test demonstration and evaluation in 3 European airports, a small, a medium sized, and a big airport, and (5) wide dissemination and elaboration of easy-to-use training material for end users. For the customs, BAG-INTEL solution aims to: increase the successful detection of contraband in luggage by at least 20%; demonstrate the possibility and utility in automatically to derive risk indicators from external data such as the Advanced Passenger Information; demonstrate the effectivity of AI camera based reidentification of luggage, when the traveller carries it into the customs space at the exit of the carousel area; increase the fluidity of passenger flow and control by at least 20%; decrease the customs personal resources mobilisation by at least 20%; derive data useful in flights risk assessment; derive data useful in flights risk assessment; demonstrate the autolearning capacity of this smart risk engine.