Using Air Traffic Control Taskload Measures and Communication Events to Predict Subjective Workload

Civil Aerospace Medical Institute2002PDFTech ReportManning, C. A.Mills, S. H.Fox, CPfleiderer, E. MMogilka, H. JUnited States. Department of Transportation. Federal Aviation Administration. Office of Aviation. Civil Aerospace Medical InstituteSBC Technology Resources, Inc.Federal Aviation Administration AcademyUnited States. Department of Transportation. Federal Aviation Administration. Office of Aviation. Office of Aerospace MedicineUnited States. Department of Transportation. Federal Aviation Administration. Office of Aviation. Civil Aerospace Medical InstituteUnited StatesAir traffic controlAir traffic controllersClearances (Navigation)CommunicationsMathematical predictionRegression analysisWorkloadTaskloadDOT/FAA/AM-02/4Final ReportA study was conducted to determine whether air traffic control (ATC) communication events would predict subjective estimates of controller workload as well as measures of controller taskload. We compared different regression models' predictions of subjective workload estimates made by 16 subject matter experts on 5 occasions during 8 samples of air traffic activity. The predictors were different combinations of four taskload principal components computed from routinely recorded ATC data, two principal components representing the number and duration of voice communication events, and two principal components representing the content of voice communications. Several regression model comparisons were computed to identify \u201creduced\u201d regression models containing fewer predictors that would predict the workload ratings as well as a full model containing all predictors. Several reduced models predicted ATWIT (Air Traffic Workload Input Technique) ratings as well as the full model but all of these contained the Activity component. These reduced models were a model containing only the Activity component, a model containing the Activity and Instructional Clearances components, and a model containing the Activity, Instructional Clearances, and All Communications Number and Duration components. The results suggest that routinely recorded ATC data provide a good estimate of subjective workload. However, if recordings of voice communications are available and researchers want to invest the time required to analyze the transcripts, they may be able to improve slightly their estimate of subjective workload. The researcher must consider whether the information gained is worth the additional time investment required for analysis.120