False alerts in the ATC conflict alert system: Is there a “cry wolf” effect?

Christopher D. Wickens; Shaun Hutchins; Stephen Rice; David Keller; Jamie S. Hughes · 2009 · Proceedings of the Human Factors and Ergonomics Society Annual Meeting

DOI: 10.1177/154193120905300120

archive: indexed pipeline: cataloged

Abstract

We analyzed the extent to which a high false alert rate of the conflict alerting (CA) system in five ATC facilities was the cause of a “cry-wolf” effect, whereby true alerts of a pending loss of separation were associated with either controller failure to respond, or a delayed response. Radar track data surrounding 497 CA's were examined and from these we extracted information as to whether the alert was true or false, whether a trajectory change was (response) or was not (non-response) evident, whether a loss of separation occurred, and the controller response time to the CA. Results revealed an overall 47% false alert rate, but that increases in this rate across facilities was not associated with more non-responses or delayed responses to true alerts, or loss-of-separation. Cry-wolf appeared to be absent. Instead, desirable anticipatory behavior indicated that controllers often responded prior to the conflict alerts.

Access

Route: Publisher paywall (check institutional access or ResearcherGate)