Effects of automation reliability in different traffic density environment on air traffic controllers conflict resolution performance

Abstract

This paper studies several hypotheses. With the help of CRA, it is postulated (H1) that there would be an improvement in the ATCOs’ performance and workload compared to carrying out the conflict resolution tasks without the aid of the CRA. Additionally, there ought to be an improvement in performance when participants experience the perfectly reliable automation with respect to the unreliable automation (H2). Nevertheless, we should still see an improvement in performance for the unreliable automation when compared with that of the manual condition where CRA is not provided (H3). The inclusion of CRA would result in lower SA, therefore it was hypothesized that participants’ SA would be reduced under CRA condition compared to a situation where CRA is not provided. Nevertheless, the SA was anticipated to be greater in the automation of imperfect reliability compared to reliable automation since the unreliability ought to rise longer processing time of raw data (H4). Through the use of ATCSimulator®2; we will develop and generate several different air traffic situations. ATCSimulator®2 will mimic the Terminal Radar Control (TRACON) facility and employ a similar separation rules as the Federal Aviation Administration (FAA). We were able to show that CRA reliability significantly affects percentage of resolved conflict. Thus we confirmed the hypothesis that performance improved with the aid of CRA. The participants were also able to respond to the SA ready question quicker under automated conditions compared to manual condition. In conclusion, the ground-based CRA concept proves to be of huge potential in resolving en route airspace capacity issues in the upcoming future. A steady shift towards more automated operations can happen as soon as additional development in research and operations have occurred.