| A myriad of workload assessment techniques are available for the estimation of air traffic controller workload. However, most of these techniques are intrusive in nature and create obstructions to their normal work. Therefore, objective of this research was to investigate if various speech parameters are affected by the air traffic controller workload. Since air traffic controllers communicate with pilots as a part of their routine activities, the voice parameters could be used for workload assessment, if proven useful. Seventeen air traffic controllers from the Atlantic City Terminal Approach Control Center participated in the study. A simulation of Atlantic City Terminal Approach Control was used for creating air traffic scenarios. Five phrases were selected for the comparison of speech parameters under various workload levels. The workload level was determined based on the instantaneous number of aircraft handled by the controller. Average and standard deviation of amplitude, average and standard deviation of fundamental frequency (Hz, and semitones), average and standard deviation of jitter and shimmer of the speech segment were the speech parameters considered as dependent variables. Controller, workload and replication were considered as independent variables. In general, under the low and high workload conditions, speech parameters did not get affected by the workload level and indicated a lack of sensitivity towards workload. However, significant individual differences among controllers were consistently observed. |
