Monitoring performance of professional and occupational operators

Abstract

The human capacity to simultaneously perform several tasks depends on the quantity and the mode of mentally processing the information imposed by the tasks. Since operational environments are highly dynamic, priorities across tasks will be expected to change as the mission evolves, thus the capability to reallocate the mental resources dynamically depending on such changes is very important. The resources required in very complex situations, such as air traffic management (ATM), can exceed the user's available resources leading to increased workload and performance impairments. In this regard, the availability of information concerning the workload experienced by the operators while dealing with tasks will be fundamental for both warning them when overload conditions are approaching and improving interactions with the system. The idea of our work was to use neurophysiologic data collected from professional air traffic controllers (ATCOs) to provide additional information to standard measures with which to assess the ATCOs' expertise and a machine learning electroencephalography-based index to evaluate their mental workload during the execution of ATC tasks. The results showed that the proposed method was able to track the workload alongside the execution of the realistic ATM scenario, and provide added values to objectively assess the expertise of the ATCOs.

Keywords: Air traffic controller; Air traffic management; EEG; Expertise; Machine learning; Mental workload; Multimodal approach; Passive brain-computer interface.