Subjective aspects of mental workload in air-traffic control

Two indices are usually used to estimate air-traffic controllers’ workload: subjective self estimation and objective estimation by counting aircraft to be monitored. However, due to uncertainty and real time pressure, the strain undergone by the controller remains difficult to evaluate objectively. The aim of this paper is to test a new index integrating both, traffic features influencing workload (monitoring, radar regulation and conflict solving) and the number of aircraft. This Demand Index (DI) was computed every ten seconds during real traffic work sessions of one hour involving 25 professional air-traffic controllers (Saint Exupéry airport, Lyon, France). Five vegetative variables (skin potential and conductance, skin blood flow and temperature and instantaneous heart rate) were recorded. Tonic level variation across time was used to evaluate activation level changes every ten seconds. After the session, each controller was asked to evaluate his own workload using the NASA TLX rating scale. If subjective self-estimation was correlated to DI, DI revealed stronger correlation with vegetative activity than the number of aircraft. Any increase in DI was closely associated with vegetative variation. Thus, emotional activity, i.e. workload subjective aspects, should be taken into account in the evaluation of workload.