Correlational analysis of survey and model-generated workload values

Abstract

This study examines the accuracy of an Army helicopter pilot workload measuring model called the Task Loading Model. The model is a submodel of the Army-NASA Aircrew/Aircraft Integration Program's Man-Machine Integration Design and Analysis System. The model's workload level output was correlated with the subjective workload measurements of several groups of pilots evaluating a variety of flight tasks. Seventy-one Army aviators completed surveys requiring scaled ratings and paired comparisons of workload related to common flight tasks conducted during typical missions. Their responses were examined for internal consistency and pooling by means of nonparametric tests. Aviator-supplied data was found to be robust and reliable. Pooled response data was correlated with model-generated data to determine the accuracy of the model. Results of this study show that the Task Loading Model is presently inadequate, but displays promising trends and should be further refined.