Human Factors Analysis of Safety Alerts in Air Traffic Control

Allendoerfer, Kenneth R; Friedman-Berg, Ferne; Pai, Shantanu · 2007 · ROSA P / United States. Department of Transportation. Federal Aviation Administration. William J. Hughes Technical Center

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This 2007 Federal Aviation Administration (FAA) technical report addresses the human factors challenges associated with safety alerts in Air Traffic Control (ATC) automation systems. The study was motivated by recent safety incidents, including a mid-air collision and a near-miss with an antenna, where controllers appeared to fail to respond appropriately to Conflict Alerts (CAs) or Minimum Safe Altitude Warnings (MSAWs). Additionally, the National Transportation Safety Board (NTSB) recommended redesigning alert systems to reliably capture controller attention. The primary research question focused on determining how and when controllers respond to these alerts, assessing the prevalence of "nuisance alerts" (alerts that are algorithmically valid but operationally unnecessary), and identifying methods to improve alert algorithms and presentations to enhance controller effectiveness and system safety. The researchers conducted a two-phase study involving field visits and data analysis. First, they observed live operations and interviewed personnel at various ATC facilities. Second, they analyzed automation data and voice recordings from five en route and 17 terminal facilities, examining 607 CAs and 178 MSAWs. The methodology involved categorizing controller responses—such as issuing traffic advisories or control instructions—and analyzing the timing of these responses relative to alert activation. The study aimed to quantify the proportion of alerts that provided useful information versus those that were redundant or resolved prior to activation. The results indicated that a significant majority of alerts were nuisance alerts. In en route facilities, 62% of CAs and 91% of MSAWs received no controller response; in terminal facilities, 44% of CAs and 61% of MSAWs received no response. Crucially, no operational errors occurred in these non-response cases. When controllers did act, they typically did so before the alert activated (67% of the time for CAs and 68% for MSAWs). Furthermore, many alerts lasted so briefly that the situation likely resolved itself or was already managed by the controller. The authors estimated that 81–87% of CAs and 87–97% of MSAWs were unnecessary. These findings suggest that high rates of nuisance alerts increase workload, desensitize controllers to genuine hazards, and reduce trust in automation. The report concludes that reducing nuisance alerts should be a top priority for the FAA. Recommendations include improving alert suppression functions, such as implementing "snooze" features that reactivate alerts only if specific criteria are met. The authors also advise basing alert algorithm parameters, like look-ahead time, on human factors data regarding controller reaction times. To mitigate the impact of remaining alerts, the study recommends graded alert presentations that increase in urgency as the situation becomes more critical. Additional suggestions include collocating speakers with displays, restricting audible volume settings to prevent disruption, and improving the salience and readability of alert data blocks. These changes aim to ensure alerts provide new, actionable information without disrupting ongoing operations.

Key finding

81-87% of Conflict Alerts and 87-97% of Minimum Safe Altitude Warnings were identified as nuisance alerts that provided no useful information to controllers.

Methodology

naturalistic

Sample size: 785

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• auditory warnings