Anzeige
Erschienen in:
26.09.2022 | Wissenschaftliche Beiträge
A time machine for accident investigators: Reconstructing the likelihood of detecting traffic during a midair collision using simulator-based experiments with flight crew
Erschienen in: Zeitschrift für Arbeitswissenschaft | Ausgabe 1/2023
EinloggenAktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
In December of 2012, two light aircraft collided near Frankfurt. The accident left investigators puzzled. All four adults in the aircraft were pilots.
Both aircraft involved were operated per visual flight rules. The crews were responsible for visually detecting and avoiding other traffic. Hampering the investigation was a lack of onboard evidence. This left investigators only with radar tracks and the wreckage as notable evidence.
The lingering question of why the collision occurred was investigated using a flight simulator to study visual attention. 25 pilot-rated test subjects participated. Researchers captured the likelihood of when conflicting traffic was visually detected as a function of the opportunity given to the participants. This allowed researchers to quantify their visual attention. The results show that literature overestimates see-and-avoid’s effectiveness in some constellations. Once established on their collision course, the accident flight crews were only offered insufficient visual cues to deconflict reliably.
Practical Relevance
On a meta-level, the investigation serves as a showcase of how tools from engineering and cognitive psychology can be combined to answer questions about sociotechnical systems. The accident investigation itself is a prime example of “Safety-I”. Its results still provide a benefit by sensitizing operators and flight crews about the potential shortcomings of the see-and-avoid principle.