Breakdowns in team resilience during aircraft landing due to mental model disconnects as identified through machine learning

Abstract

Landing system in aviation is a representative sociotechnical system; such systems often require seamless teamwork among experts. Resilience is critical for the ability of a team to adapt to sudden changes, overcome challenges, and return to normalcy. However, team heterogeneity can result in cognitive misalignment, known as mental model disconnects (MMDs). MMDs can cause interaction conflcits, which compromise team resilience. This study aimed to use machine learning to identify breakdowns in team resilience during aircraft landing. The results showed that the MMDs in the terminal area had the greatest influence on missed approaches; in particular, the late intervention of air traffic controllers’ (ATCOs) in an improper landing state was the major breakdown identified. This result highlights the importance of ATCO initiative in preventing landing incidents. The mechanism was determined to be ineffective information dissemination in nonroutine scenarios, which unravels the necessity for improving future information systems.